Method for quantitative marketing of telemarketing clue data, storage medium, electronic equipment and product

Abstract

The invention relates to the technical field of business marketing, and particularly provides a telemarketing clue data quantitative marketing method, a storage medium, electronic equipment and a product. The telemarketing clue data quantitative marketing method comprises the following steps: determining a daily target issuing amount of various levels of clue data in outbound clue data and an issuing freezing period of various types of clue data under the various levels of clue data; issuing a daily clue data packet based on the daily target issuing amount of the various levels of clue data and the issuing freezing period of the various types of clue data under the various levels of clue data; and after the outbound task corresponding to the clue data packet is completed, recalling clue data to be revisited, and distributing the clue data to be revisited to a manual service agent. According to some embodiments of the invention, the marketing quality and the marketing conversion rate can be improved while the user experience is not affected.

Description

Technical Field

[0001] The present application relates to the field of telephone marketing technology, and in particular to a method, storage medium, electronic device and product for quantitative marketing of telephone marketing lead data. Background Art

[0002] With the continuous development of intelligent customer service, it has been widely used in different business fields.

[0003] In order to reduce labor costs, currently in the telemarketing scenario, we mainly rely on intelligent customer service to make outbound calls to promote products to users. At present, the number of outbound leads that intelligent customer service can handle per unit time is limited. If we do not pre-screen and grade the quality of lead data and perform differentiated cycle operations, making indiscriminate outbound calls for massive lead data will result in the loss of high-intent users and a low business marketing conversion rate.

[0004] Therefore, how to provide a technical solution for a method of quantitative marketing of telemarketing lead data with a higher conversion rate has become a technical problem that needs to be solved urgently. Summary of the invention

[0005] The purpose of some embodiments of the present application is to provide a method, storage medium, electronic device and product for quantitative marketing of telemarketing lead data. Through the technical solutions of the embodiments of the present application, the conversion rate and marketing quality of business marketing can be improved without affecting the user experience.

[0006] In the first aspect, some embodiments of the present application provide a method for quantitative marketing of telemarketing lead data, including: determining the daily target issuance volume of various levels of lead data in the outbound lead data and the issuance freezing period of various types of lead data under the various levels of lead data; issuing daily lead data packets based on the daily target issuance volume of the various levels of lead data and the issuance freezing period of various types of lead data under the various levels of lead data; after the outbound call task corresponding to the lead data packet is completed, recalling the lead data to be followed up, and assigning the lead data to be followed up to manual agents.

[0007] Some embodiments of the present application determine the daily target delivery volume of lead data of various levels and the delivery freezing period of each type of lead data, and then deliver the daily lead data package. Then, after the outbound call task is completed, the lead data to be followed up can be recalled and assigned to manual agents for service. In this way, on-demand delivery of different types of lead data can be achieved, thereby improving the conversion rate and marketing quality of business marketing without affecting the user experience.

[0008] In some embodiments, determining the daily target sending volume of various levels of lead data in the outbound lead data and the sending freezing period of each type of lead data under the various levels of lead data includes: classifying and cross-classifying the outbound lead data to obtain multiple levels of lead data; determining the daily target sending volume of various levels of lead data among the multiple levels of lead data; and determining the sending freezing period of each type of lead data among the multiple types of lead data in the daily target sending volume of the various levels of lead data based on the reach status, wherein the reach status of each type of lead data is different.

[0009] Some embodiments of the present application classify and cross-classify the outbound call lead data and determine the daily target delivery volume of lead data of various levels, and then determine the delivery freeze period of various lead data according to the contact status. Different delivery periods are formulated for different types of lead data, and outbound calls are made to related users according to the delivery periods, which can improve user experience and marketing quality.

[0010] In some embodiments, the outbound call lead data is classified and cross-classified to obtain multiple levels of lead data, including: using a quantitative model to perform multi-dimensional quantitative classification on the outbound call lead data to obtain multi-dimensional lead data; and cross-processing each dimensional lead data in the multi-dimensional lead data to obtain the multiple levels of lead data.

[0011] Some embodiments of the present application perform quantitative classification and cross-processing on outbound call lead data through a quantitative model to obtain lead data of various levels. Through data level classification, data of different qualities are distinguished by level to improve marketing quality.

[0012] In some embodiments, determining the daily target delivery volume of clue data of various levels among the multiple levels of clue data includes: obtaining the data volume and delivery cycle of clue data of various levels; and taking the ratio of the data volume and delivery cycle of clue data of various levels as the daily target delivery volume of clue data of various levels.

[0013] Some embodiments of the present application determine the daily target delivery volume of clue data of various levels through the data volume and delivery cycle of clue data of various levels, so as to achieve balanced delivery of clue data.

[0014] In some embodiments, the contact status includes: an outbound call not connected status, a connected but not transferred to manual status, and a connected and transferred to manual status.

[0015] In some embodiments, the determination of the delivery freeze period of each type of clue data in the multiple types of clue data in the daily target delivery volume of the clue data of various levels based on the reach status includes: obtaining the proportion of data in different reach statuses in the clue data of various levels; dividing the clue data of various levels into the multiple types of clue data based on the data proportion; determining the delivery freeze period of each type of clue data through parameter search, inferring each type of available clue data according to the delivery freeze period, and compiling the daily target delivery volume of clue data of various levels according to each type of available clue data under clue data of various levels.

[0016] Some embodiments of the present application can achieve targeted delivery of lead data and improve marketing quality by determining the delivery freeze period under different reach states of lead data at various levels.

[0017] In some embodiments, the recalling of the lead data to be revisited includes: using preset rules to recall the lead data to be revisited from a preset outbound call task; wherein the preset rules include: at least one of a recall lead model, a data recall level strategy, and a keyword tag screening strategy.

[0018] Some embodiments of the present application recall the data of leads to be revisited according to preset rules so as to facilitate revisits by human agents, thereby improving the marketing conversion rate.

[0019] In a second aspect, some embodiments of the present application provide a device for quantitative marketing of telemarketing lead data, including: a determination module for determining the daily target issuance volume of lead data of various levels in the outbound lead data and the issuance freezing period of each type of lead data under the various levels of lead data; a issuance module for issuing daily lead data packets based on the daily target issuance volume of lead data of various levels and the issuance freezing period of each type of lead data under the various levels of lead data; a recall module for recalling the lead data to be returned after the outbound call task corresponding to the lead data packet is completed, and allocating the lead data to be returned to a manual agent.

[0020] In a third aspect, some embodiments of the present application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can implement the method described in any embodiment of the first aspect.

[0021] In a fourth aspect, some embodiments of the present application provide an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement a method as described in any embodiment of the first aspect when executing the program.

[0022] In a fifth aspect, some embodiments of the present application provide a computer program product, wherein the computer program product comprises a computer program, wherein the computer program, when executed by a processor, can implement the method described in any embodiment of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of some embodiments of the present application, the drawings required for use in some embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0024] Figure 1 A system diagram of quantitative marketing of telemarketing lead data provided for some embodiments of the present application;

[0025] Figure 2 One of the flow charts of the method for quantitative marketing of telemarketing lead data provided in some embodiments of the present application;

[0026] Figure 3 A schematic diagram of clue classification cross-cutting results provided for some embodiments of the present application;

[0027] Figure 4 Flowchart 2 of the method for quantitative marketing of telemarketing lead data provided for some embodiments of the present application;

[0028] Figure 5 A block diagram of the apparatus for quantitative marketing of telemarketing lead data provided in some embodiments of the present application;

[0029] Figure 6 A schematic diagram of an electronic device is provided for some embodiments of the present application. DETAILED DESCRIPTION

[0030] The technical solutions in some embodiments of the present application will be described below in conjunction with the drawings in some embodiments of the present application.

[0031] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0032] In the relevant technology, the amount of outbound leads that intelligent customer service can handle in a unit of time is limited. If pre-screening and grading of the lead data quality and differentiated periodic operations are not performed, indiscriminate outbound calls will be made to massive amounts of lead data, high-intent users will be lost, and the business marketing conversion rate will be low. For users who do not need the product, this will not only waste marketing resources, but also cause user disgust.

[0033] In view of this, some embodiments of the present application provide a method for quantitative marketing of telemarketing lead data, which is mainly aimed at telemarketing scenarios that combine man and machine. The method classifies the lead data that can be called outbound, determines the freezing period for issuing different types of lead data, and then after the daily outbound call task is completed, the lead data to be returned can be recalled in the remaining time and assigned to manual seats for active return service. The embodiments of the present application can issue lead data by category and period, and can conduct targeted marketing to users, thereby improving marketing quality and marketing conversion rate.

[0034] The following is combined with Figure 1 The overall composition structure of the system for quantitative marketing of telemarketing lead data provided by some embodiments of the present application is exemplified.

[0035] like Figure 1 As shown, some embodiments of the present application provide a system for quantitative marketing of telemarketing lead data, which may include: a terminal 100 and a server 200. The terminal 100 may transfer the outbound call lead data to the server 200, and the server 200 may analyze and process the outbound call lead data, determine the daily target delivery volume of lead data of various levels and the delivery freeze period of various lead data, and determine the lead data package delivered daily based on this. After the outbound call task is completed, the lead data to be returned can also be recalled and assigned to a manual agent for return service.

[0036] In some embodiments of the present application, the terminal 100 may be a mobile terminal or a non-portable computer terminal, which is not specifically limited in the embodiments of the present application.

[0037] In other embodiments of the present application, if the terminal 100 has the function of analyzing, processing, issuing, recalling and distributing the outbound callable clue data, the server 200 may not be provided. The specific selection may be made according to the actual situation, and the embodiments of the present application are not limited thereto.

[0038] The following is combined with Figure 2 The implementation process of the quantitative marketing of telemarketing lead data performed by the server 200 provided in some embodiments of the present application is exemplified.

[0039] Please see attached Figure 2 , Figure 2A flow chart of a method for quantitative marketing of telemarketing lead data provided for some embodiments of the present application, the method for quantitative marketing of telemarketing lead data may include:

[0040] S210, determining the daily target delivery volume of various levels of lead data in the outbound callable lead data and the delivery freezing period of various types of lead data under various levels of lead data.

[0041] For example, in some embodiments of the present application, a clue is a case, which is a sample in a machine learning model. For example, in the fields of insurance and the like, a user and the user-related information are generally called a clue, and the clue data is some basic information related to the user. Since users have different needs and activity levels, the clue data can be classified to determine the freezing period for the issuance of various types of clue data.

[0042] In some embodiments of the present application, S210 may include:

[0043] S211, classify and cross-classify the outbound callable clue data to obtain clue data of multiple levels.

[0044] For example, in some embodiments of the present application, the outbound call lead data is first classified, and different outbound call dialing strategies can be subsequently formulated through the classification.

[0045] In some embodiments of the present application, S211 may include: using a quantization model to perform multi-dimensional quantization on the outbound callable clue data to obtain multi-dimensional clue data; cross-processing each dimension clue data in the multi-dimensional clue data to obtain the multiple levels of clue data.

[0046] For example, in some embodiments of the present application, the entire amount of outbound callable lead data is classified and modeled by a single multi-objective quantitative model, multiple single-objective quantitative models, etc. Modeling is performed based on the characteristics X such as the qualifications and intentions of the lead data and the historical communication labels y, and a machine learning model model (as a specific example of a quantitative model) with circle selection and stratification capabilities is trained. Among them, the qualifications of the leads can be understood as the quality level of the user's own attributes; the intention of the leads can be understood as the user's willingness to purchase the product.

[0047] For example, taking the qualifications and intentions of leads as an example, multiple single objectives are selected to model the qualifications and intentions of the lead data respectively, and the "lead qualification quantification model model1" (as a specific example of the quantification model) and the "lead intention quantification model model2" (as a specific example of the quantification model) are obtained through training. The full amount of outbound call lead data of the estimated level is input into model1 and model2 respectively to obtain the qualification and intention level labels of each lead (as a specific example of multi-dimensional lead data), and then these two types of data are cross-processed to obtain the final level label of each lead. Accordingly, the full amount of outbound call lead data will be classified under various level labels to obtain multiple levels of lead data.

[0048] For example, Figure 3 As shown in the schematic diagram of the cross-results, the clue qualification quantification model outputs three types of clue stratification: qualification top 1 clue class, qualification top 2 clue class, qualification top 3 clue class; the clue intention quantification model outputs two types of clue stratification: intention top 1 clue class, intention top 2 clue class. The clue classifications of the above two dimensions are crossed to obtain six types of clue data. These six types of clue data carry level labels respectively. Level labels may include: high-quality clue data, low-quality clue data and other level labels. It can be understood that in addition to the above-mentioned quantification dimensions of clue qualifications and intentions, other dimensions can be added according to actual application scenarios; level labels can also be divided more finely according to actual application scenarios (for example, higher-quality clue data, medium-quality clue data, etc.). It should be understood that the embodiments of the present application are not limited to this.

[0049] S212, determining a daily target delivery volume of clue data of each level among the multiple levels of clue data.

[0050] For example, in some embodiments of the present application, after the clue data level division is completed, the daily target delivery volume for each level can be determined.

[0051] In some embodiments of the present application, S212 may include: obtaining the data volume and delivery cycle of clue data of various levels; and using the ratio of the data volume and delivery cycle of clue data of various levels as the daily target delivery volume of clue data of various levels.

[0052] For example, in some embodiments of the present application, clues of various levels are first issued in a mixed ratio, and the strategy for issuing clues of various levels in a mixed ratio is that the issuance interval (as a specific example of the issuance cycle) is different for clue data of different levels. For example, the issuance interval of high-quality clue data is short, and the issuance interval of low-quality clue data is long. In this way, while ensuring the quality of daily operational clues, the full ecological operation of the full amount of outbound callable clue data can be taken into account to avoid the Matthew effect, that is, the worse the quality of the clues, the less chance there is to add new communication information, the less the model knows about the characteristics of these clues, and the less likely it is to have an opportunity to improve the evaluation of these clues. After that, the daily target issuance volume is determined by the amount of clues (that is, the amount of data) and the issuance interval of the clue data.

[0053] For example, Figure 3 Take the clue data of for example, let's note that "Qualification top 1 & intention top 1 clue category" is high-quality clue data A, and the number of clues at this level is 20w; "Qualification top 3 & intention top 2 clue category" is low-quality clue data B, and the number of clues at this level is 30w. According to the strategy of "mixed distribution of clues at each level in proportion", the distribution interval of 20w high-quality clue data A is set to 4 days, that is, about 5w clue data of this level are distributed every day (as a specific example of the number of clues distributed); the distribution interval of 30w low-quality clue data B is set to 10 days, that is, about 3w clue data of this level are distributed every day.

[0054] S213, based on the reaching status, determining the freezing period for the delivery of each type of clue data in the multiple types of clue data in the daily target delivery volume of the clue data of various levels, wherein the reaching status of each type of clue data is different. The reaching status includes: outbound call not connected status, connected but not transferred to manual status, and connected transferred to manual status.

[0055] For example, in some embodiments of the present application, according to the strategy of configuring differentiated sending cycles for each category of clues according to the outbound call reach depth, a sending freeze cycle is determined for each category of clue data. Among them, the overall idea of ​​the strategy of "configuring differentiated sending cycles for each category of clues according to the outbound call reach depth" is that "the deeper the reach depth of the clue data, the longer the interval between the days of sending the clue data, and the less likely it will trigger the user's negative emotions; when the reach depth of the clue data is shallow, the interval between the days of sending can be appropriately reduced, and the business revenue indicators can be taken into account." Here, the reach depth refers to the different response states of the user when calling the user corresponding to the clue data, usually including user failure to connect (i.e., outbound call failure), user connection but not transferred to a person (i.e., connection but not transferred to a manual state, only communicating with the AI ​​robot), user connection and transfer to a person (i.e., connection and transfer to a manual state), etc. Obviously, user failure to connect is a shallower level of contact, user connection but not transferred to a person belongs to the middle level of contact, and user connection and transfer to a person is a deeper level of contact. In the system, the relationship between the reach status and the interval number of days between dispatches (as a specific example of a dispatch freeze period) can be determined through parameter search, so that the interval number of days between dispatches for various types of clue data can be quickly determined.

[0056] In some embodiments of the present application, S213 may include: obtaining the proportion of data in different reach states in the various levels of clue data; based on the data proportion, dividing the various levels of clue data into the multiple categories of clue data; determining the issuance freeze period of each type of clue data through parameter search, inferring each type of available clue data according to the issuance freeze period, and compiling the daily target issuance volume of clue data of various levels according to each type of available clue data under various levels of clue data.

[0057] For example, in some embodiments of the present application, based on the data proportion of clue data under different trigger states in the previously issued clue data, multiple categories of clue data are obtained through the data proportion, and then the number of days between issuance of each category of clue data is determined through parameter search.

[0058] For example, taking the daily issuance of 50,000 high-quality clue data as an example, after the outbound call is made, assuming that 80% of the users are not connected (80%*5w=4w), 14% of the users are connected but not transferred (14%*5w=0.7w), and 6% of the users are connected but transferred (6%*5w=0.3w), then the issuance cycle days for 80% of the users who are not connected (that is, the issuance interval days) becomes 2 days, the issuance cycle days for 14% of the users who are connected but not transferred becomes 7 days, and the issuance cycle days for 6% of the users who are connected and transferred becomes 14 days. Specifically, the issuance interval days corresponding to the current contact state can be determined from the comparison table, and marked on this type of clue data. When issuing it later, it can be issued according to the issuance cycle days. It should be noted that the way to change the issuance cycle days of low-quality clue data is the same as the processing method of the above-mentioned high-quality clue data, and the clue data of other levels are also the same. In order to avoid repetition, it will not be repeated here.

[0059] It should be noted that through the cross-action of the two strategies proposed above, before sending out 50,000 data of high-quality lead data A every day, it will be pre-calculated whether each lead in the 200,000 high-quality lead data A has reached the next outbound call sending cycle (i.e., the lead is unfrozen). If it has reached the sending cycle, it will be sent; if not, it cannot be sent.

[0060] S220, based on the daily target delivery volume of the clue data of various levels and the delivery freezing period of various types of online data under the clue data of various levels, deliver the daily clue data package.

[0061] For example, in some embodiments of the present application, according to the sending cycle of each type of clue data, the clue data that can be sent on the same day is selected from the outbound call clue data. If the total amount of a certain type of clue data that can be sent on the same day is less than the daily target sending amount of this type of clue, it can be supplemented by the available clue data of other categories at this level or other levels. For example, high-quality clue data needs to be sent 5w, but some clue data has not reached the sending cycle. At this time, it can be supplemented with clue data of other categories to make up a clue data package that meets the target quantity requirements.

[0062] S230, after the outbound call task corresponding to the clue data packet is completed, the clue data to be returned is recalled, and the clue data to be returned is allocated to a manual agent.

[0063] For example, in some embodiments of the present application, when the task of quantitatively sending clue data packets daily (as a specific example of an outbound call task) is consumed, the manual seat will actively make outbound calls based on the list of "multi-channel recall clues" (as the clue data to be returned). For example, if a manual seat's daily demand is 120 clue data (adjustable as needed), and 100 clues have been completed on the day, 20 of the list of "multi-channel recall clues" can be allocated to the manual seat.

[0064] In some embodiments of the present application, S230 may include: using preset rules to recall the lead data to be revisited from a preset outbound call task; wherein the preset rules include: at least one of a recall lead model, a data recall level strategy, and a keyword tag screening strategy.

[0065] For example, in some embodiments of the present application, a list of "multi-channel recall leads" can be obtained according to different preset rules. One or more preset rules can be used for recall, and the embodiments of the present application are not specifically limited here. After the recall is carried out according to the above preset rules, the multi-channel recall lead data can be merged according to the snake merge algorithm, so as to ensure the stability of the daily outbound call connection rate and the transfer rate, and obtain a high-quality list of "multi-channel recall leads".

[0066] For example, based on the recall of data analysis rules (as a specific example of data recall level strategy), the lead data missed by the transferee of the previous day's outbound call task (as a specific example of a preset outbound call task) is prioritized for recall; or, based on the conversation content of the previous outbound call cycle (as a specific example of a preset outbound call task), LLM (as a specific example of a recall lead model) is used to extract user qualifications and intention tags, and the lead data corresponding to the conversation content with a high-intention tag set is prioritized for recall; or, based on the conversation content of the previous outbound call cycle, the high-intention keyword vocabulary is hit (as a specific example of a keyword tag screening strategy), and priority is given to recall.

[0067] Although three recall strategies are provided here, the recall strategies can be flexibly adjusted in actual applications to recall high-quality lead data. The embodiments of the present application are not limited to this. Through multi-channel recall of lead data, some high-quality lead data can be selected, allowing artificial agents to refine operations and improve the business conversion indicators of this part of the lead data.

[0068] The following is combined with Figure 4 The specific process of quantitative marketing of telemarketing lead data provided by some embodiments of the present application is exemplified.

[0069] Please see attached Figure 4 , Figure 4A flow chart of a method for quantitative marketing of telemarketing lead data is provided for some embodiments of the present application.

[0070] The above process is explained below as an example.

[0071] S410, using a quantitative model to perform multi-dimensional quantitative classification on the outbound callable lead data to obtain multi-dimensional lead data.

[0072] S420, cross-processing each dimension clue data in the multi-dimensional clue data to obtain clue data of multiple levels.

[0073] S430, determining a daily target delivery volume of clue data of each level among the clue data of multiple levels.

[0074] S440, based on the reach status, determining a freezing period for issuing various types of clue data among multiple types of clue data in a daily target issuing volume of clue data of various levels.

[0075] S450, determining clue data packets to be issued daily based on the daily target issuance volumes of clue data of various levels and the issuance freezing periods of various clue data.

[0076] S460, confirming that the outbound call task corresponding to the clue data packet is completed.

[0077] S470, using preset rules, recalling the lead data to be returned from the preset outbound call task.

[0078] S480, allocating the data of the leads to be returned to the human agents for active return visit service.

[0079] It should be noted that the specific implementation process of S410 to S480 can refer to the method embodiment provided above, and in order to avoid repetition, the detailed description is appropriately omitted here.

[0080] Through some of the above embodiments of the present application, it can be seen that the embodiments of the present application are based on the cooperation between the two links of quantitatively sending clue data and multi-channel recall clues, which can improve the operational efficiency of clue data. Among them, quantitatively sending clue data can improve data utilization efficiency, ensure the quality of daily outbound call data, and stabilize indicators such as connection rate and transfer rate. Multi-channel recall clues can pre-identify high-quality clues through LLM models and other rule strategies, reach high-intent users through active outbound calls, and improve marketing conversion rates.

[0081] Please refer to Figure 5 , Figure 5The block diagram of the composition of the device for quantitative marketing of telemarketing lead data provided by some embodiments of the present application is shown. It should be understood that the device for quantitative marketing of telemarketing lead data corresponds to the above method embodiment and can execute each step involved in the above method embodiment. The specific functions of the device for quantitative marketing of telemarketing lead data can be found in the description above. To avoid repetition, the detailed description is appropriately omitted here.

[0082] Figure 5 The device for quantitative marketing of telemarketing lead data includes at least one software function module that can be stored in a memory in the form of software or firmware or solidified in the device for quantitative marketing of telemarketing lead data, and the device for quantitative marketing of telemarketing lead data includes: a determination module 510, used to determine the daily target issuance volume of lead data of various levels in the outbound call lead data and the issuance freezing period of each type of lead data under the various levels of lead data; a issuance module 520, used to issue daily lead data packets based on the daily target issuance volume of lead data of various levels and the issuance freezing period of each type of lead data under the various levels of lead data; a recall module 530, used to recall the lead data to be returned after the outbound call task corresponding to the lead data packet is completed, and assign the lead data to be returned to a manual agent.

[0083] In some embodiments of the present application, a determination module 510 is used to classify and cross-classify the outbound callable lead data to obtain multiple levels of lead data; determine the daily target delivery volume of each level of lead data among the multiple levels of lead data; and determine the delivery freeze period of each type of lead data among the multiple types of lead data in the daily target delivery volume of the various levels of lead data based on the reach status, wherein the reach status of each type of lead data is different.

[0084] In some embodiments of the present application, the determination module 510 is used to perform multi-dimensional quantitative classification on the outbound callable clue data using a quantitative model to obtain multi-dimensional clue data; and cross-process each dimension clue data in the multi-dimensional clue data to obtain the multiple levels of clue data.

[0085] In some embodiments of the present application, the determination module 510 is used to obtain the data volume and the issuance cycle of clue data of various levels; and the ratio of the data volume and the issuance cycle of clue data of various levels is used as the daily target issuance volume of clue data of various levels.

[0086] In some embodiments of the present application, a determination module 510 is used to obtain the data proportions in different reach states in the clue data of various levels; based on the data proportions, the clue data of various levels are divided into the multiple categories of clue data; the sending freeze period of each category of clue data is determined through parameter search, each category of available clue data is inferred according to the sending freeze period, and the daily target sending volume of clue data of various levels is gathered according to each category of available clue data under clue data of various levels.

[0087] In some embodiments of the present application, the recall module 530 is used to recall the lead data to be revisited from the preset outbound call task using preset rules; wherein the preset rules include: at least one of a recall lead model, a data recall level strategy, and a keyword tag screening strategy.

[0088] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the device described above can refer to the corresponding process in the aforementioned method, and will not be described in detail here.

[0089] Some embodiments of the present application further provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can implement the operations of the method corresponding to any of the above methods provided in the above embodiments.

[0090] Some embodiments of the present application further provide a computer program product, which includes a computer program, wherein when the computer program is executed by a processor, it can implement the operations corresponding to any of the above methods provided in the above embodiments.

[0091] like Figure 6 As shown, some embodiments of the present application provide an electronic device 600, which includes: a memory 610, a processor 620, and a computer program stored in the memory 610 and executable on the processor 620, wherein the processor 620 can implement a method as described in any of the above embodiments when reading the program from the memory 610 through a bus 630 and executing the program.

[0092] Processor 620 can process digital signals and can include various computing structures, such as complex instruction set computer structure, reduced instruction set computer structure, or a structure that implements a combination of multiple instruction sets. In some examples, processor 620 can be a microprocessor.

[0093] The memory 610 may be used to store instructions executed by the processor 620 or data related to the execution of instructions. These instructions and / or data may include codes for implementing some or all functions of one or more modules described in the embodiments of the present application. The processor 620 of the disclosed embodiment may be used to execute instructions in the memory 610 to implement the method shown above. The memory 610 includes a dynamic random access memory, a static random access memory, a flash memory, an optical memory, or other memory known to those skilled in the art.

[0094] The above description is only an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.

[0095] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0096] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

Claims

1. A method for quantitative marketing of telemarketing lead data, characterized in that: include: Determine the daily target delivery volume of various levels of lead data in the outbound callable lead data and the delivery freeze period of various types of lead data under various levels of lead data; Based on the daily target delivery volume of clue data of various levels and the delivery freezing period of various clue data under the clue data of various levels, delivering daily clue data packets; After the outbound call task corresponding to the clue data packet is completed, the clue data to be returned is recalled and the clue data to be returned is allocated to a manual agent.

2. The method according to claim 1, characterized in that The determination of the daily target delivery volume of various levels of lead data in the outbound callable lead data and the delivery freezing period of various types of lead data under various levels of lead data includes: Classify and cross-classify the outbound callable clue data to obtain clue data of multiple levels; Determine a daily target delivery volume of clue data of various levels among the multiple levels of clue data; Based on the reaching status, the sending freezing period of each type of clue data in the multiple types of clue data in the daily target sending volume of the clue data of various levels is determined, wherein the reaching status of each type of clue data is different.

3. The method according to claim 2, characterized in that The outbound callable clue data is classified and cross-classified to obtain clue data of multiple levels, including: Using a quantitative model to perform multi-dimensional quantitative classification on the outbound callable clue data to obtain multi-dimensional clue data; Cross-processing is performed on each dimension clue data in the multi-dimensional clue data to obtain the multiple levels of clue data.

4. The method according to claim 2 or 3, characterized in that The determining of the daily target delivery volume of clue data of various levels among the clue data of multiple levels includes: Obtain the data volume and delivery cycle of clue data at various levels; The ratio of the data volume of clue data of various levels to the sending cycle is used as the daily target sending volume of clue data of various levels.

5. The method according to claim 2 or 3, characterized in that: The contact status includes: an outbound call not connected status, a connected but not transferred to manual status, and a connected and transferred to manual status.

6. The method according to claim 5, characterized in that The step of determining the freezing period for sending each type of clue data in the multiple types of clue data in the daily target sending amount of clue data of various levels based on the reaching status includes: Obtaining the proportion of data in different reaching states among the clue data of various levels; Based on the data proportions, the various levels of clue data are divided into the multiple categories of clue data; The freezing period for issuing various types of clue data is determined through parameter search, each type of available clue data is inferred based on the freezing period for issuing, and the daily target issuing volume of clue data of various levels is gathered based on each type of available clue data under clue data of various levels.

7. The method according to any one of claims 1 to 3 and 6, characterized in that: The recalled lead data to be revisited includes: The preset rules are used to recall the lead data to be returned from the preset outbound call tasks; wherein the preset rules include: at least one of a recall lead model, a data recall level strategy, and a keyword tag screening strategy.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program executes the method according to any one of claims 1 to 7 when executed by a processor.

9. An electronic device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the computer program executes the method according to any one of claims 1 to 7 when being run by the processor.

10. A computer program product, characterized in that The computer program product comprises a computer program, wherein the computer program executes the method according to any one of claims 1 to 7 when executed by a processor.

Images