Customer maintenance method, device and computer program product

Abstract

The invention relates to the technical field of customer management, in particular to a customer maintenance method and device and a computer program product. The customer maintenance method comprises the following steps: acquiring an information field in customer information, and determining an information category of the information field; determining a basic score of the information field, and determining a category score of each information category based on the basic score of the information field included in each information category; adjusting the weight corresponding to each information category according to the customer interaction feedback information, and performing weighted summation on the category score of each information category to determine the comprehensive score of the customer; and determining a maintenance action on the customer according to the comprehensive score.

Description

Technical Field

[0001] This disclosure relates to the field of customer management technology, and in particular to a customer maintenance method, apparatus and computer program product. Background Technology

[0002] Currently, business departments within companies are responsible for managing customer relationships, which includes maintaining those relationships through maintenance activities.

[0003] Because customer data is scattered across different functional modules and lacks systematic integration, it is impossible to comprehensively depict customer needs. This results in maintenance actions that do not match the actual needs of customers, leading to wasted resources and difficulty in accurately reaching target customers through maintenance actions, thus hindering business implementation. Summary of the Invention

[0004] To overcome the problems existing in the related technologies, this disclosure provides a customer maintenance method, apparatus and computer program product that can solve the above problems.

[0005] According to a first aspect of the present disclosure, a customer maintenance method is provided, the method comprising: Collect information fields from customer information and determine the information categories of the information fields; determine the base score of the information fields, and determine the category score of each information category based on the base scores of the information fields contained in each information category; adjust the weights corresponding to each information category according to customer interaction feedback information, and perform a weighted summation of the category scores of each information category to determine the customer's comprehensive score; determine the maintenance actions for the customer based on the comprehensive score.

[0006] According to a second aspect of the present disclosure, a customer maintenance apparatus is provided, the apparatus comprising: The data collection unit is configured to collect information fields from customer information and determine the information categories of the information fields; the first determination unit is configured to determine the base score of the information fields and determine the category score of each information category based on the base scores of the information fields contained in each information category; the weighting unit is configured to adjust the weights corresponding to each information category according to customer interaction feedback information and perform a weighted summation of the category scores of each information category to determine the customer's comprehensive score; the second determination unit is configured to determine the maintenance actions for the customer based on the comprehensive score.

[0007] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a processor and a memory; the memory being used to store a computer program; and the processor being used to execute the customer maintenance method as described in the first aspect by invoking the computer program.

[0008] According to a fourth aspect of the present disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method as described in the first aspect.

[0009] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects: This disclosure allows for the collection of customer information, the classification of information fields within that information, the determination of corresponding information categories, and the determination of base scores for each information field. Furthermore, it allows for the determination of category scores for each information category based on customer feedback regarding recent maintenance actions. The weights of each information category are then adjusted, and a weighted sum of the category scores for each information category is calculated to determine the customer's overall score. Finally, maintenance actions for the customer are determined based on the overall score.

[0010] The method disclosed herein categorizes information fields, determining the score of customer information within multiple categories based on these categories, and using these scores to reflect the customer's actual needs. Furthermore, this disclosure allows for adjustment of weights based on customer interaction feedback, enabling a more accurate determination of the customer's overall score using both recorded routine customer information and interactive feedback. This more accurately reflects customer needs and allows for more precise execution of appropriate maintenance actions for customers with different requirements. On one hand, it avoids or reduces maintenance costs associated with performing maintenance actions on customers without business needs; on the other hand, performing appropriate maintenance actions on more precisely targeted customers is more conducive to achieving business objectives.

[0011] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0012] The accompanying drawings, which are incorporated in and form part of this disclosure, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0013] Figure 1 This is a schematic flowchart illustrating a customer maintenance method according to an exemplary embodiment of the present disclosure.

[0014] Figure 2 This is a block diagram illustrating a customer maintenance device according to an exemplary embodiment of the present disclosure.

[0015] Figure 3 This disclosure is a schematic block diagram illustrating a customer maintenance device according to an exemplary embodiment. Detailed Implementation

[0016] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0017] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The singular forms “a,” “the,” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0018] It should be understood that although the terms first, second, third, etc., may be used in this disclosure to describe various information, such information should not be limited to these terms. These terms are used only to distinguish information of the same type from one another. For example, without departing from the scope of this disclosure, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0019] In industries such as finance and retail, which require precise customer maintenance, business departments need to maintain customer relationships. Due to the diversification of business modules (such as customer management, business management, and event management), customer-related data can be stored in multiple functional modules, which makes the relevant technologies have some shortcomings in acquiring customer data for customer maintenance.

[0020] First, the lack of direct data exchange between multiple functional modules has exacerbated the problem of data silos. For example, basic customer information (such as customer level and customer size) can be stored in the customer management module; behavioral data (such as call response rate and SMS open rate) can be stored in the call management module; and business interaction data (such as event attendance rate and maintenance item collection records) can be stored in the event management module. This lack of systematic integration between multiple independent modules prevents a comprehensive understanding of customer needs, resulting in a problem of "data fragmentation."

[0021] Secondly, the related technologies are relatively simplistic in their data usage. Most of these technologies use single-dimensional data (i.e., customer information obtained from a single functional module) to identify customer tags, or fuse multi-dimensional data through fixed weight ratios. They do not take into account the dynamic changes of customers, resulting in data results that do not match the actual needs of customers. This makes it difficult to accurately perform effective maintenance actions for specific customer groups, thus causing them to miss the window of opportunity for customers' business needs.

[0022] Furthermore, the related technologies lack feedback and cannot correct generated tags. After generating customer tags, these technologies only store the customer tags as static results. Static results may expire, making the effectiveness of maintenance actions performed based on expired customer tags unpredictable. This could lead to the blind deployment of maintenance resources and waste of resources.

[0023] Finally, the related technologies suffer from poor adaptability and high costs. The multi-dimensional information fusion solutions employed by these technologies require modifications to existing business systems to enable data interaction between multi-functional modules before data fusion can be achieved. This modification process is costly, time-consuming, and difficult to adapt to the enterprise's existing business systems (customer management architecture).

[0024] In summary, the relevant technologies result in inefficient customer maintenance, low resource utilization, and an inability to meet enterprises' needs for accurate, efficient, and low-cost customer maintenance.

[0025] To address the aforementioned technical issues, this disclosure proposes a customer maintenance method.

[0026] Figure 1 This is a schematic flowchart illustrating a customer maintenance method according to an embodiment of the present disclosure, which can be executed by a processing device.

[0027] like Figure 1 As shown, customer maintenance methods include: In step S101, information fields from customer information are collected, and the information category of the information fields is determined; In step S102, the basic score of the information field is determined, and the category score of each information category is determined based on the basic score of the information field contained in each information category; In step S103, the weights corresponding to each information category are adjusted based on customer interaction feedback information, and the category scores of each information category are weighted and summed to determine the customer's overall score. In step S104, maintenance actions for the customer are determined based on the comprehensive score.

[0028] In some embodiments, information fields in customer information are collected, and the information category of the information fields is determined.

[0029] Customer information can be obtained through various channels and methods, and then information fields can be identified within that information. For customer information stored across multiple platforms and modules, corresponding information retrieval methods for each platform and module can be employed.

[0030] For example, the data input module can be connected to the functional modules of an existing data platform to obtain customer information stored in those modules. These functional modules may include, but are not limited to, customer management, wealth management, merchant management, call management, SMS management, lobby functions, outreach management, event management, and maintenance item distribution and verification modules.

[0031] Information fields can be pre-defined fields used to characterize customer traits. After determining customer information, the information fields within that information can be further defined to reflect the customer's characteristics.

[0032] The information category of an information field can be determined based on its type and / or the category of customer characteristics it represents. In one embodiment, semantic analysis can be performed on the information fields based on semantic feature recognition, thereby clustering them into similar information categories. In another embodiment, the information categories can be preset and correspond to the information fields.

[0033] In some embodiments, a base score for the information field is determined, and a category score for each information category is determined based on the base scores of the information fields contained in each information category.

[0034] The base score for an information field can be determined. This base score characterizes a customer's score in that field, and the score helps determine the customer's category for accurate customer segmentation and maintenance.

[0035] An information category contains information fields that are closely related to one direction of customer management. An information category can contain multiple information fields, and the category score is determined based on the base scores of these multiple information fields.

[0036] It should be noted that the more information fields a category contains, the more characteristic points of the customer in that management direction it can reflect, and the more accurate the determined category score will be. This disclosure does not limit the specific method for determining the category score based on the base score; for example, it can be an arithmetic sum or an arithmetic average.

[0037] In some embodiments, the weights corresponding to each information category are adjusted based on customer interaction feedback information, and the category scores of each information category are weighted and summed to determine the customer's overall score.

[0038] By combining the category scores of multiple information categories to determine the overall score, the actual needs of customers can be identified more accurately.

[0039] However, in some scenarios, the recorded customer information is updated slowly, and the customer's recent maintenance actions and corresponding customer feedback may not be recorded in the customer information, nor are they formed into effective information fields.

[0040] Therefore, we can further determine the customer's interactive feedback information regarding recent maintenance actions, and then adjust the weights corresponding to each information category based on the customer's interactive feedback information, and determine the comprehensive score based on the adjusted weights.

[0041] Since interactive feedback information is based on recent maintenance actions, it can more accurately reflect the customer's current intentions and needs compared to static information fields. As a result, the overall score determined by the adjusted weights can more accurately reflect the customer's characteristics and recent tendencies, which is conducive to determining more accurate maintenance actions.

[0042] It should be noted that adjusting the weights corresponding to each information category based on customer interaction feedback and then summing the category scores of each information category to determine the customer's overall score is only one specific method. Based on this example, it is relatively simple to determine customer characteristics by combining recent interaction feedback information with customer information. In some embodiments, the score corresponding to the interaction feedback information can also be determined, and then the overall score can be determined based on the interaction feedback information score and the category score.

[0043] In some embodiments, maintenance actions for customers are determined based on the overall score.

[0044] Maintenance actions for customers can be determined based on their overall scores. The overall score indicates the customer's maintenance value; customers with higher overall scores can be prioritized for maintenance, or partial maintenance actions can be determined based on the overall score.

[0045] For example, maintenance actions may include inviting customers to meetings and / or distributing maintenance gifts. Based on the methods disclosed herein, customers with a comprehensive score higher than a first score can be invited to attend offline meetings, while customers with a comprehensive score not lower than a second score can be offered maintenance gifts. Customers with a low comprehensive score can be re-engaged through low-cost maintenance methods such as telephone and SMS.

[0046] This disclosed customer maintenance method collects information fields from customer information and determines the corresponding information categories. By determining the base score for each information field, a category score for each information category is determined. Weights are then adjusted based on customer feedback regarding maintenance actions, thus merging the category scores of multiple information categories to determine the customer's comprehensive score. Based on this comprehensive score, specific maintenance actions are determined for that customer. This method determines the category scores representing multiple aspects of each information category and adjusts the weights based on interactive feedback. This comprehensive score, incorporating diverse customer information and recent, more timely feedback, provides a more complete picture of customer needs. After accurately identifying the customer, maintenance actions are implemented based on the comprehensive score, making maintenance more efficient and targeted. These actions can be more precisely triggered to facilitate business transactions and avoid wasting maintenance costs.

[0047] In some embodiments, determining the information category of the information field includes: determining the information field as an information category based on the relevance of the customer maintenance needs represented by the information field; the information category includes at least one of the following: basic customer information, customer stickiness information, and business interaction information.

[0048] The information category corresponding to an information field can be determined based on the relevance of the customer's maintenance needs. An information category can contain multiple information fields, and these multiple information fields represent a high degree of relevance to the customer's maintenance needs.

[0049] For example, multiple information categories can be preset, and the category of customer maintenance needs corresponding to each information category can be determined; when information fields are determined, information fields can be classified into corresponding information categories based on the relevance of customer maintenance needs represented by the information fields.

[0050] For example, multiple information categories and corresponding information fields for each category can be preset. When determining information fields from customer information, only the information fields that correspond to the information categories can be identified, and then the identified information fields can be assigned to the corresponding information categories.

[0051] In some embodiments, the information categories include at least one of the following: basic customer information, customer loyalty information, and business interaction information.

[0052] There are three preset information categories: basic customer information, customer loyalty information, and business interaction information.

[0053] The customer basic information includes fields that characterize the customer's basic information, such as, but not limited to, asset size, financial preferences, customer level, and merchant type. For example, the corresponding information field could be "Asset size 800,000".

[0054] Customer stickiness information includes fields that characterize customer behavior data, such as, but not limited to, call response rate, SMS campaign link open rate, frequency of in-store visits, and duration of online interaction. For example, a relevant information field could be "call response rate 54%".

[0055] The business interaction information includes fields that characterize the content of the interaction with the customer, such as, but not limited to, the number of outreach follow-ups, participation in business gatherings, maintenance item collection records, and activity conversion rates. For example, the corresponding information field could be "3 outreach follow-ups".

[0056] In some embodiments, collecting information fields from customer information includes: determining the information category of the information field to be collected; and determining the collection method based on the information category of the information field to be collected.

[0057] In some technical scenarios, customer information fields can be stored on multiple platforms (functional modules). These platforms have certain functional divisions, so the information fields stored on a single platform are usually of the same information category.

[0058] Therefore, before collecting information fields, it is advisable to first determine the information category of the fields to be collected, and then determine the collection method based on the determined information category. Collection can be carried out from the platform corresponding to that information category, or it can be carried out according to the collection frequency corresponding to that information category.

[0059] For example, taking the above embodiment as an example, since the information fields in the customer basic information category are updated less frequently, the collection frequency can be determined to be once a day. The information fields in this category (e.g., asset size, financial preferences, customer level, merchant type) are usually stored in the customer management module, wealth management module, and merchant management module. Therefore, information fields can be collected from the corresponding functional modules according to the collection frequency of this information category. The collection frequency for customer stickiness information and business interaction information can be set to real-time collection, meaning that when the information fields recorded in the corresponding functional modules are updated, the updated information fields are obtained synchronously. Customer stickiness information fields (e.g., call response rate, SMS activity link open rate, lobby visit frequency, online interaction duration) can be obtained from the call management module, SMS management module, and lobby functional module; while business interaction information (e.g., number of outreach follow-ups, business gathering participation, maintenance item collection records, activity conversion rate) can be obtained from the outreach management module, activity management module, and maintenance item distribution and verification module.

[0060] In some embodiments, the collected information fields may be preprocessed.

[0061] After collecting the information fields, the raw data can be preprocessed, such as cleaning and desensitization, to eliminate invalid data and privacy risks, which will facilitate the subsequent determination of the basic score and calculation.

[0062] For example, preprocessing may include data cleaning, which can remove invalid data collected. Invalid data may include, but is not limited to, null values, outliers (e.g., “SMS activity link open rate > 100%”, but this data cannot exceed 100%), and duplicate data (e.g., duplicate outreach records for the same customer).

[0063] Preprocessing may also include filling in missing fields, which is used to fill in data where key fields are missing.

[0064] The missing asset size of another client can be filled using the "mean fill" method. For example, the missing asset size of another client can be filled using the average asset value of other clients of the same level.

[0065] Alternatively, the "historical data extension" method can be used to fill in the missing visit frequency data for the most recent time period. For example, the average data within a historical time period can be used to fill in the missing visit frequency data for the hall in the most recent time period.

[0066] Preprocessing can also include data anonymization. Irreversible encryption can be used to replace fields in a customer's private data (such as name, phone number, original asset value, etc.), thus achieving data anonymization. For example, the name "Li Si" becomes the anonymized data "Li*" after replacement, and the phone number becomes the anonymized data "139****3000" after replacement. The anonymized data complies with data security standards, reducing the risk of data leakage.

[0067] In some embodiments, a base score for the information field is determined.

[0068] Semantic modeling can be used to determine the content of information fields and assign a base score based on the ranking of that content within the same information field. Within the same information field (e.g., asset size), a higher ranking corresponds to a larger base score, indicating higher customer value. In this example, the base score can be determined based on the customer's information field's ranking percentage within the same information field.

[0069] For example, if Customer A's assets are 800,000 yuan and they rank in the top 20% of all customers' assets, then Customer A's base score for their assets could be in the top 20%. If the preset base score is 5 points, then Customer A's base score for their assets would be 4 points.

[0070] Alternatively, information fields can be standardized and mapped based on preset standards. Based on these standards, the corresponding scores for each information field can be determined relatively easily and quickly.

[0071] For example, the asset size field in the information field can be divided into 5 levels based on the data range: <100,000 has a base score of 1 point, 100,000-500,000 has a base score of 2 points, 500,000-2,000,000 has a base score of 3 points, 2,000,000-5,000,000 has a base score of 4 points, and >5,000,000 has a base score of 5 points. As another example, the SMS link open rate can also be divided into 5 levels: <20%, 20%-40%, 40%-60%, 60%-80%, and >80%, corresponding to 1-5 points respectively. The financial preference content in the information field can be text, with text content categorized as conservative, stable, balanced, growth-oriented, and aggressive, corresponding to base scores of 2, 3, 4, 4.5, and 5 points respectively.

[0072] In some embodiments, determining the category score of each information category based on the base scores of the information fields contained in each information category includes: determining the arithmetic mean of the base scores of the information fields contained in each information category as the category score.

[0073] The information fields are standardized and normalized by using basic scores, and then the standardized data is fused at the feature layer to determine the category score of the information field corresponding to the information category.

[0074] We can determine the base score for each information field within each information category, and then calculate the arithmetic mean of these base scores to obtain the category score for that information category. The expression can be as follows: Category score = (basic score of information field 1 + basic score of information field 2 + ... + basic score of information field n) / n.

[0075] For example, the information fields and corresponding base scores for the information category "Customer Basic Information" are: Asset Size 3 points, Financial Preference 3 points, Customer Level 4 points. Therefore, the category score for this information category (Customer Basic Information) is (3+3+4) / 3 = 3.33 points.

[0076] The score can be rounded to two decimal places for easier calculation and display.

[0077] In some embodiments, after determining the category scores for each information category, the category scores for multiple information categories can be merged to determine the customer's overall score.

[0078] For example, a base weight can be preset for each information category, with the sum of the base weights being 1. This ensures that the total category score of all information categories does not exceed the upper limit of the base score.

[0079] For example, information categories can be assigned basic weights based on their degree of influence on customer intentions. As an example, the weight of basic customer information could be 0.4, the weight of customer stickiness information could be 0.3, and the weight of business interaction information could be 0.3.

[0080] The customer's overall score is determined by weighted summation of the category scores of multiple information categories based on weights.

[0081] However, the overall score is determined based on the acquired customer information, which may be somewhat outdated and may not reflect recent customer interactions with maintenance requests. Therefore, as a better implementation, the base weights can be adjusted based on customer interaction feedback, so that the overall score can reflect customer interaction feedback.

[0082] In some embodiments, adjusting the weight corresponding to each information category based on customer interaction feedback information includes: presetting a basic weight for each information category; wherein the sum of the basic weights is 1; determining customer interaction feedback information in response to historical maintenance actions; determining the adjustment weight of the information category corresponding to the customer interaction feedback information, and adjusting the basic weight of the corresponding information category.

[0083] The adjustment coefficient can be determined based on customer interaction feedback, and then the basic weight of the information category can be adjusted based on the adjustment coefficient. The range of the adjustment coefficient can be determined as [-0.1, 0.1].

[0084] Customer interaction feedback can be based on recent maintenance actions (e.g., the last 3 times or the last 2 weeks). For example, if a customer registers and attends an invited business gathering, it indicates that the customer may have a real need for the business in the near future, and the corresponding adjustment factor could be +0.1; or, based on collected customer information, if the call response rate is determined to be less than or equal to 30% for two consecutive times, it indicates that the customer's willingness has declined, and the corresponding adjustment factor could be -0.05; or, if a customer's asset size increases by more than or equal to 20% within 30 days, it indicates that the customer may have potential business needs, and the corresponding adjustment factor could be +0.05.

[0085] On the other hand, the corresponding information categories can be determined based on customer interaction feedback. For example, if a customer registers for and attends a business gathering, it corresponds to business interaction information; if the call response rate is less than or equal to 30% for two consecutive times, it corresponds to customer loyalty information; if the asset size grows by more than or equal to 20% within 30 days, it corresponds to basic customer information.

[0086] The base weight of an information category can be adjusted based on an adjustment coefficient determined from customer interaction feedback information to determine the adjustment weight. Adjustment weight = Base weight × (1 + Adjustment coefficient).

[0087] It should be noted that in this disclosure, the information field is a static depiction of the customer's status, while the interactive feedback information is a dynamic verification result of the customer's status in response to maintenance actions. The information field and the interactive feedback information differ fundamentally in nature and function; therefore, they correspond to different stages in determining the overall score. The information field is used to determine the category score, while the interactive feedback information is used to adjust the weights used to determine the overall score.

[0088] In some embodiments, the category scores for each information category are weighted and summed to determine the customer's overall score.

[0089] The overall score is obtained by summing the product of the category score for each information category and the adjusted weight for that information category.

[0090] For example, if the category scores for the customer's three information categories are 3.33, 2.67, and 3.67 respectively, and the adjusted weights are [0.396, 0.312, 0.232], then the overall score = 3.33 × 0.396 + 2.67 × 0.312 + 3.67 × 0.232 ≈ 3.25 points.

[0091] In some embodiments, the method further includes: determining the comprehensive score as a first score if the comprehensive score is less than a first score; and / or determining the comprehensive score as a second score if the comprehensive score is greater than a second score.

[0092] Because the weights are adjusted based on the basic weights for each information category, the combined score determined by the adjusted basic weights of the three information categories may not fall within a certain range (e.g., not within the 1-5 range). In this case, the combined score can be limited to the appropriate range.

[0093] If the total score is less than the first score (e.g., 1 point), the total score will be set as the first score. If the total score is greater than the second score (e.g., 5 points), the total score will be set as the second score.

[0094] Based on this approach, on the one hand, it ensures that interactive feedback information can more accurately and truthfully reflect (verify) customer situations, without being restricted to the rule that the sum of weights is 1, and without needing to increase the weight of one information category while correspondingly decreasing the weight of another; on the other hand, it ensures that the overall score is within a reasonable range, facilitating data processing.

[0095] The following example will explain in detail how to determine the overall score.

[0096] The method disclosed herein can transform multidimensional information into a comprehensive customer score, which can be mainly divided into two stages: feature-level fusion and decision-level fusion.

[0097] Feature layer fusion refers to converting the standardized score of a single information field into a category score for an information category. It can transform multiple information fields of the same information category (such as the information category "customer basic data", which includes asset size, financial preferences, etc.) into a unified "category score", eliminating the dimensional differences between multiple information fields within the same category.

[0098] The preprocessed standardized scores (1-5 points) can be grouped according to information categories (e.g., basic customer information, customer stickiness information, business interaction information). Then, the arithmetic mean of all information fields under each information category is calculated, and the result is rounded to two decimal places to ensure accuracy, serving as the category score. If any information field is missing data, it can be filled in during the preprocessing stage, thus not affecting the calculation.

[0099] Determine the category score for each of the three information categories. For example, basic customer information = 3.33 points, customer stickiness information = 2.67 points, and business interaction information = 3.67 points.

[0100] Decision-level integration refers to the process of converting category scores into a comprehensive customer score. This comprehensive score reflects the customer's actual maintenance needs, and its weight dynamically changes based on customer feedback.

[0101] The base weights (0.4, 0.3, 0.3) for the three information categories can be adjusted based on feedback data from the customer's last three maintenance actions (e.g., participation in financial planning sessions, duration of maintenance item redemption, etc.). A feedback coefficient can be determined based on the feedback data, and then the base weights can be adjusted accordingly. Finally, a weighted sum is calculated based on the adjusted weights to determine the overall score. The overall score can be limited to a range of 1-5 points, where a score higher than 5 points is counted as 5 points, and a score lower than 1 point is counted as 1 point. For example, a customer's overall score could be 3.30 points.

[0102] The feedback coefficient is used to dynamically adjust the base weights. The range of the feedback coefficient can be [-0.1, 0.1], which can avoid excessive fluctuations in weights that could distort the overall score.

[0103] The rules for determining feedback coefficients based on feedback data can be found in the following table: Table of Correspondence between Feedback Data, Information Categories, and Feedback Coefficients

[0104] In some embodiments, the method further includes: determining a static label based on the customer's category score, the static label being used to characterize the customer's features.

[0105] Static tags can be determined based on category scores of basic customer information and are updated quarterly. They can be used to reflect long-term stable customer attributes and characterize customer features.

[0106] For example, if the category score of the customer's basic information is greater than 4, the corresponding static label is "high-net-worth customer" (in the wealth management field, this could be an aggressive wealth management customer); if the category score is between 2 and 4, the corresponding static label is "medium-asset customer" (in the wealth management field, this could be a conservative wealth management customer); if the category score indicates that the customer is a retail merchant, the static label will also be the corresponding retail merchant.

[0107] In some embodiments, corresponding business products can be pushed to customers with specific static tags, or corresponding maintenance actions can be taken.

[0108] Once customers have their static tags identified, these tags can be used to roughly differentiate customers. Customers with the same static tag can be approached with similar maintenance actions and similar service products can be recommended. Specific customer groups have different actual needs for different service products, so the more suitable service products can be recommended to these groups to facilitate a sale.

[0109] For example, taking wealth management products as an example, higher-yield products can be pushed to clients with the static label "aggressive wealth management client," and maintenance actions for clients with this label can also be product-oriented. For "conservative wealth management clients," the product yields pushed may be lower, but the risk level is lower, and the corresponding maintenance actions should focus on building client trust.

[0110] In some embodiments, the method further includes: determining dynamic tags based on the customer's overall score and the customer interaction feedback information, the dynamic tags being used to characterize the customer's maintenance needs.

[0111] Dynamic tags can reflect customers' real-time maintenance needs and can be updated at intervals (e.g., 7 days).

[0112] Dynamic tags can be determined based on a customer's overall score and interactive feedback. Matching the overall score with the interactive feedback information allows for the precise identification of high-quality customers through dynamic tags, facilitating subsequent maintenance and follow-up; it can also identify inactive customers, allowing for the development of specific follow-up plans for them.

[0113] For example, if the overall score is greater than 4 points and the interactive feedback information indicates a high level of participation in wealth management, the corresponding dynamic tag could be "high-potential wealth management customer, priority for wealth management promotion"; if the overall score is between 3 and 4 points and the feedback information includes that maintenance products have not been redeemed after being received, the corresponding dynamic tag could be "customer who needs maintenance products to activate, key follow-up in external expansion"; if the overall score is less than 3 points and the feedback information indicates that the frequency of visits to the lobby is less than 2 times per month, the corresponding dynamic tag could be "dormant customer, priority for SMS wake-up".

[0114] Customer tags can be stored in a preset format for easy retrieval. The preset format can be "tag code + tag name + generation time", for example, "001-High-potential financial customer-2025XXXX".

[0115] In some embodiments, determining the maintenance action for the customer based on the comprehensive score includes: determining the maintenance action for the customer based on the comprehensive score and the customer's corresponding tag.

[0116] Maintenance actions can be performed in a targeted manner based on comprehensive scores, static tags, and dynamic tags.

[0117] Maintenance actions may include, but are not limited to, sending links, business product notifications, SMS reminders, and phone calls. Alternatively, after determining that offline maintenance is necessary and providing complimentary maintenance supplies, technical personnel may perform the maintenance actions.

[0118] In some embodiments, the method further includes: determining customer feedback data in response to the maintenance action and a tag corresponding to the customer; adjusting the weight of the information category in the tag corresponding to the feedback data if the feedback data meets preset conditions, and / or adjusting the comprehensive score threshold of the tag.

[0119] This disclosure can also optimize the process of determining the comprehensive score through closed-loop optimization.

[0120] Feedback data (feedback optimization information) can be collected, and optimizations can be determined based on this data. Feedback data mainly includes: activity matching feedback, maintenance item write-off feedback, and asset change feedback.

[0121] Activity matching feedback refers to determining whether customers with active dynamic tags and high perceived demand have responded to activities, registered for activities, or completed transactions.

[0122] Maintenance item verification feedback refers to the verification time for customers who are generally active but whose needs are unclear after the maintenance action of issuing maintenance items is performed. It can be divided into three levels: less than 7 days, 7-15 days, and more than 15 days.

[0123] Asset change feedback refers to the rate of change in a customer's asset size within 30 days after the static and / or dynamic tag is pushed. It can also be divided into three levels: growth, no change, and decline.

[0124] If the feedback data meets the preset conditions, feedback adjustments are made.

[0125] If the feedback data meets the preset conditions but does not meet expectations, it indicates that the labels determined based on the comprehensive score are inaccurate, and / or the comprehensive score is difficult to accurately characterize the customer. Therefore, the comprehensive score can be adjusted so that the adjusted comprehensive score can more accurately characterize the customer, and the maintenance actions taken for the customer can achieve the expected results.

[0126] For example, if the activity matching feedback for a dynamic tag representing a relatively active customer indicates an activity conversion rate of less than 10% (e.g., the conversion rate for "high-potential financial clients" is 8%, less than 10%), the base weight of the corresponding information category for that dynamic tag can be reduced. If the tag is primarily determined based on basic customer information, then the base weight of that basic customer information can be reduced by 5%, for example, from 0.4 to 0.35.

[0127] For example, if the maintenance product redemption feedback for dynamically tagged, generally active customers shows a redemption rate greater than 80% (e.g., 85% for "customers needing maintenance products to activate"), this indicates excessive redemptions and high maintenance costs. It's possible that some customers with good intentions who don't need maintenance products are being underestimated and have had their redemptions made. In this case, the overall score threshold for that tag can be lowered, adjusting the range of overall scores for that tag to adjust the corresponding customer group. For instance, if the original overall score range for "customers needing maintenance products to activate" was 3-4 points, excessive redemptions suggest that the needs and intentions of customers with scores of 3-4 are being underestimated. Therefore, the overall score threshold can be lowered to 2.8-3.8 points for that tag, allowing maintenance products to be distributed to customers who truly need them while reducing costs.

[0128] The dynamic weight algorithm can be retrained based on the full amount of feedback data at preset intervals (e.g., 30 days) to update the rules for taking the values ​​of the feedback coefficients, thereby continuously optimizing the model and more accurately determining the customer's comprehensive score, identifying the customer's tags, and taking more accurate maintenance actions for the customer.

[0129] Corresponding to the embodiments of the customer maintenance method of this disclosure, this disclosure also provides embodiments of a corresponding customer maintenance device.

[0130] Please see Figure 2 , Figure 2This is a block diagram of a customer maintenance device according to one embodiment of this disclosure. Figure 2 As shown, the customer maintenance device includes: The acquisition unit 210 is configured to acquire information fields from customer information and determine the information category of the information fields; The first determining unit 220 is configured to determine the basic score of the information field and determine the category score of each information category based on the basic score of the information field contained in each information category; The weighting unit 230 is configured to adjust the weights corresponding to each information category based on customer interaction feedback information, and to perform a weighted summation of the category scores of each information category to determine the customer's overall score. The second determining unit 240 is configured to determine maintenance actions for the customer based on the comprehensive score.

[0131] In some embodiments, determining the information category of the information field includes: determining the information field as an information category based on the relevance of the customer maintenance needs represented by the information field; the information category includes at least one of the following: basic customer information, customer stickiness information, and business interaction information.

[0132] In some embodiments, determining the category score of each information category based on the base scores of the information fields contained in each information category includes: determining the arithmetic mean of the base scores of the information fields contained in each information category as the category score.

[0133] In some embodiments, adjusting the weight corresponding to each information category based on customer interaction feedback information includes: presetting a basic weight for each information category; determining customer interaction feedback information in response to historical maintenance actions; determining the adjustment weight of the information category corresponding to the customer interaction feedback information; and adjusting the basic weight of the corresponding information category.

[0134] In some embodiments, the apparatus is further configured to: determine the comprehensive score as a first score if the comprehensive score is less than a first score; and / or determine the comprehensive score as a second score if the comprehensive score is greater than a second score.

[0135] In some embodiments, the apparatus is further configured to determine a static label based on the customer's category score, the static label being used to characterize the customer's features.

[0136] In some embodiments, the device is further configured to determine dynamic tags based on the customer's overall score and the customer interaction feedback information, the dynamic tags being used to characterize the customer's maintenance needs.

[0137] In some embodiments, the device is further configured to: determine customer feedback data in response to the maintenance action and a tag corresponding to the customer; and, if the feedback data meets preset conditions, adjust the weight of the information category in the tag corresponding to the feedback data, and / or adjust the comprehensive score threshold of the tag.

[0138] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0139] Embodiments of this disclosure also provide an electronic device, including: a processor and a memory; the memory for storing a computer program; and the processor for executing a customer maintenance method as described in any of the above embodiments by invoking the computer program.

[0140] Embodiments of this disclosure also provide a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the customer maintenance method as described in any of the foregoing embodiments.

[0141] Embodiments of this disclosure also provide a computer program product, including a computer program that, when executed by a processor, implements the method as described in any of the foregoing embodiments.

[0142] Figure 3 This is a schematic block diagram illustrating a customer maintenance device 300 according to embodiments of the present disclosure. For example, device 300 may be a mobile phone, computer, digital broadcast terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.

[0143] Reference Figure 3 The device 300 may include one or more of the following components: processing component 302, memory 304, power supply component 306, multimedia component 308, audio component 310, input / output (I / O) interface 312, sensor component 314, and communication component 316.

[0144] Processing component 302 typically controls the overall operation of device 300, such as operations associated with display, telephone calls, data communication, camera operation, and recording. Processing component 302 may include one or more processors 320 to execute instructions to complete all or part of the steps of the aforementioned customer maintenance method. Furthermore, processing component 302 may include one or more modules to facilitate interaction between processing component 302 and other components. For example, processing component 302 may include a multimedia module to facilitate interaction between multimedia component 308 and processing component 302.

[0145] Memory 304 is configured to store various types of data to support operation of device 300. Examples of such data include instructions for any application or method operating on device 300, contact data, phonebook data, messages, pictures, videos, etc. Memory 304 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0146] Power supply component 306 provides power to various components of device 300. Power supply component 306 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to device 300.

[0147] Multimedia component 308 includes a screen that provides an output interface between the device 300 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 308 includes a front-facing camera and / or a rear-facing camera. When the device 300 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0148] Audio component 310 is configured to output and / or input audio signals. For example, audio component 310 includes a microphone (MIC) configured to receive external audio signals when device 300 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 304 or transmitted via communication component 316. In some embodiments, audio component 310 also includes a speaker for outputting audio signals.

[0149] I / O interface 312 provides an interface between processing component 302 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, start buttons, and lock buttons.

[0150] Sensor assembly 314 includes one or more sensors for providing status assessments of various aspects of device 300. For example, sensor assembly 314 may detect the on / off state of device 300, the relative positioning of components such as the display and keypad of device 300, changes in the position of device 300 or a component of device 300, the presence or absence of user contact with device 300, the orientation or acceleration / deceleration of device 300, and temperature changes of device 300. Sensor assembly 314 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 314 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 314 may also include an accelerometer, a gyroscope, a magnetometer, a pressure sensor, or a temperature sensor.

[0151] Communication component 316 is configured to facilitate wired or wireless communication between device 300 and other devices. Device 300 can access wireless networks based on communication standards, such as WiFi, 2G, 3G, 4G LTE, 5G NR, or combinations thereof. In one exemplary embodiment, communication component 316 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 316 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0152] In an exemplary embodiment, the device 300 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the customer maintenance method described above.

[0153] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 304 including instructions, which can be executed by a processor 320 of the device 300 to complete the aforementioned customer maintenance method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0154] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0155] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

[0156] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0157] The methods and apparatus provided in the embodiments of this disclosure have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this disclosure. The descriptions of the embodiments above are only for the purpose of helping to understand the methods and core ideas of this disclosure. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this disclosure. Therefore, the content of this specification should not be construed as a limitation of this disclosure.

Claims

1. A customer maintenance method, characterized in that, The method includes: Collect information fields from customer information and determine the information category of the information fields; Determine the base score for the information field, and determine the category score for each information category based on the base score of the information fields contained in each information category; The weights corresponding to each information category are adjusted based on customer interaction feedback, and the category scores of each information category are weighted and summed to determine the customer's overall score. The maintenance actions for the customer are determined based on the overall score.

2. The method according to claim 1, characterized in that, Determining the information category of the information field includes: The information fields are determined as information categories based on their relevance to the customer maintenance needs they represent. The information categories include at least one of the following: basic customer information, customer loyalty information, and business interaction information.

3. The method according to claim 1, characterized in that, The determination of the category score for each information category based on the base scores of the information fields contained in each information category includes: The arithmetic mean of the base scores of the information fields contained in each information category is determined as the category score.

4. The method according to claim 1, characterized in that, The step of adjusting the weight corresponding to each information category based on customer interaction feedback includes: Pre-set basic weights for each information category; Determine customer interaction feedback information regarding customer responses to historical maintenance actions; Determine the adjustment weight of the information category corresponding to the customer interaction feedback information, and adjust the base weight of the corresponding information category.

5. The method according to claim 4, characterized in that, The method further includes: If the overall score is less than the first score, the overall score will be set as the first score; and / or, If the overall score is greater than the second score, the overall score will be determined as the second score.

6. The method according to claim 1, characterized in that, The method further includes: Static labels are determined based on the customer's category score, and these static labels are used to characterize the customer's features.

7. The method according to claim 1, characterized in that, The method further includes: Dynamic tags are determined based on the customer's overall score and customer interaction feedback information. These dynamic tags are used to characterize the customer's maintenance needs.

8. The method according to any one of claims 6-7, characterized in that, The method further includes: Determine the customer's feedback data regarding the maintenance actions and the corresponding tags for the customer; If the feedback data meets the preset conditions, adjust the weight of the information category corresponding to the feedback data in the label, and / or adjust the comprehensive score threshold of the label.

9. A customer maintenance device, characterized in that, The device includes: The data collection unit is configured to collect information fields from customer information and determine the information category of the information fields; The first determining unit is configured to determine the basic score of the information field and determine the category score of each information category based on the basic score of the information field contained in each information category; The weighting unit is configured to adjust the weights corresponding to each information category based on customer interaction feedback information, and to perform a weighted summation of the category scores of each information category to determine the customer's overall score; The second determining unit is configured to determine maintenance actions for the customer based on the comprehensive score.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the customer maintenance method as described in any one of claims 1 to 8.

Images