Multi-channel customer consultation intelligent distribution and response system for bank outsourcing
By designing an intelligent triage and response system with modules such as multi-channel access, hierarchical triage, and encrypted storage, the system solves the problems of chaotic information transmission, inaccurate triage, and data security in bank outsourcing consultations, and achieves efficient, secure, and flexible consultation processing and response.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HENAN ZHONGJIN INFORMATION TECHNOLOGY CO LTD
- Filing Date
- 2026-03-19
- Publication Date
- 2026-06-23
Smart Images
Figure CN122268982A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of financial services technology, and in particular to a multi-channel intelligent customer inquiry triage and response system for bank outsourcing. Background Technology
[0002] As the scale of banking business expands and customer inquiries continue to rise, most banks outsource routine customer inquiries to control operating costs and improve service efficiency. Currently, bank customer inquiries are handled through various channels including telephone, online banking, mobile banking, WeChat official accounts, and SMS. However, existing outsourced inquiries service models have many shortcomings and fail to meet the demands of bank service standards and customer experience.
[0003] In practical applications, existing equipment suffers from fragmented multi-channel consultation management. Different channels require separate coordination between the bank and outsourcing teams, and inconsistent communication protocols and data formats across channels lead to issues such as information transmission chaos and missed consultation requests, increasing the channel coordination and maintenance costs for both parties. Secondly, the lack of unified consultation triage rules, relying heavily on manual allocation, fails to accurately match outsourced agents based on channel priority, business type, and urgency. This results in uneven workloads for some agents, delayed responses to high-urgency inquiries, and excessively long customer wait times. Furthermore, the standardization of outsourced services and data security are difficult to guarantee. Outsourced personnel have varying skill levels and lack standardized response criteria, leading to inconsistent replies and incorrect information. Additionally, customer inquiries often contain sensitive information such as account numbers and ID numbers, and the existing model cannot effectively isolate the access permissions of outsourced personnel, posing a risk of sensitive information leakage. Moreover, adjusting triage rules and response content requires modifying the core system code, resulting in poor flexibility and an inability to quickly adapt to bank business updates and outsourcing team adjustments, leading to high maintenance costs and hindering practical application and operation. Summary of the Invention
[0004] One objective of this invention is to provide a multi-channel intelligent customer inquiry triage and response system for bank outsourcing.
[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: a multi-channel customer consultation intelligent triage and response system for bank outsourcing, including a multi-channel access module, a consultation preprocessing module, a hierarchical triage module, an outsourcing response module, a rule management module, and an encrypted storage module, wherein the modules interact with each other through an internal bus;
[0006] The multi-channel access module is used to simultaneously access consultation requests from telephone voice, online banking, mobile banking, WeChat official accounts and SMS channels, parse the protocol format of each channel and convert it into a unified data structure;
[0007] The consultation preprocessing module is connected to the multi-channel access module and performs noise filtering and keyword extraction on consultation requests. The keyword extraction is based on a preset financial business dictionary for precise matching.
[0008] The hierarchical triage module is connected to the consultation preprocessing module and the rule management module respectively. It calls the preset triage rule library to prioritize and assign seats to consultation requests. The triage rule library includes three-dimensional judgment logic of channel weight, business type and urgency.
[0009] The outsourced response module connects to the hierarchical distribution module, pushes consultation information and response templates with desensitized labels to the outsourced agent terminals, receives agent feedback and synchronizes it to the bank's core system;
[0010] The rule management module is used by bank administrators to configure traffic dictionaries, update business dictionaries and response templates, and supports setting permission isolation based on the outsourced team dimension.
[0011] The encrypted storage module performs symmetric encryption storage on consultation records, diversion logs, and anonymized customer data, and only grants complete data query permissions to authorized bank terminals.
[0012] Preferably, the multi-channel access module includes a protocol adaptation sub-module, which presets parsing rules for more than twenty mainstream channel communication protocols, and supports expanding the adaptation capability through configuration files when adding new channels without modifying the core code.
[0013] Preferably, the noise filtering process of the consultation preprocessing module is as follows: first, remove duplicate characters and meaningless auxiliary words; then, filter valid consultation content through a character length threshold; and finally, mark high-priority requests containing keywords such as complaint, emergency, and freeze.
[0014] Preferably, the three-dimensional determination logic of the hierarchical diversion module is as follows:
[0015] Channel weighting: Telephone channel weight (1.2) > Mobile banking channel weight (1.0) > WeChat and online banking channel weight (0.8) > SMS channel weight (0.6);
[0016] Business type dimension: Match agent business tags in the order of account anomaly, credit consultation, wealth management business, and regular inquiry;
[0017] Urgency level: Three priority levels are set based on keyword matching results, with high priority requests having a response time limit of ≤15 seconds.
[0018] Preferably, the outsourced response module includes a template calling submodule. The response template adopts a fixed text and variable placeholder structure. The variable placeholders are automatically filled with de-identified data obtained from the bank's core system. The de-identified data masks the customer's complete account number, ID number, and transaction amount details.
[0019] Preferably, the process for obtaining the de-identified data is as follows: the outsourced response module sends a query request with a session identifier to the bank's core system, the core system verifies the validity of the session and returns the de-identified data, and the data transmission is encrypted using the TLS1.3 protocol.
[0020] Preferably, the rule management module includes a version control unit, which retains historical versions of modifications to the traffic diversion rules and response templates, supports rollback to the three most recent valid configurations, and all modifications must be approved by two bank personnel before taking effect.
[0021] Preferably, the hierarchical traffic distribution module also includes a load balancing unit, which collects the current consultation processing volume and historical response time of outsourced agents in real time. When the processing volume of a single agent exceeds a preset threshold, it automatically distributes new requests to low-load agents with the same business tag.
[0022] Preferably, the outsourced response module also includes an exception handling unit, which automatically triggers a manual escalation process when the consultation request cannot match the keywords in the business dictionary, transferring the request to the bank's internal customer service and synchronizing the collected consultation context data.
[0023] Preferably, the encrypted storage module adopts a timed backup mechanism, performs incremental backups every day at midnight, retains backup data for more than 90 days, and the backup files can only be restored after being verified by the bank's Ukey.
[0024] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0025] (1) This invention eliminates the need for banks and outsourcing teams to separately connect to different consultation channels. The system uniformly accesses and converts data formats, reducing channel connection costs. The three-dimensional hierarchical routing rules can quickly assign consultations to suitable agents. Combined with load balancing, it avoids uneven workloads for some agents. High-priority requests are responded to within 15 seconds, shortening customer waiting time and improving consultation processing efficiency. Through permission isolation and desensitized template responses, outsourcing agents can quickly obtain standardized response content without having to memorize a large number of complex business details, reducing outsourcing personnel training costs. At the same time, it strictly restricts data access permissions for outsourcing agents, preventing the leakage of sensitive customer information and addressing banks' concerns about outsourcing data security. Meanwhile, the rules can be flexibly configured, supporting the addition of new channels and the updating of business types to adapt to the needs of bank outsourcing team adjustments and business expansion.
[0026] (2) In this invention, all modifications to the routing rules and response templates require double review, the operation is traceable and supports version rollback, which meets the compliance requirements of banking and financial business; data transmission adopts an encryption protocol, storage is encrypted throughout and backed up regularly, which can prevent data loss and avoid leakage of sensitive information, thus building a solid data security defense line; through precise routing, the probability of outsourced agents handling cross-business inquiries is reduced, and the time spent on ineffective work is reduced; response templates can be directly reused, reducing the time for agents to write responses and increasing the processing volume per unit time; it does not rely on complex technology, but operates based on preset rules, with low maintenance costs, and is suitable for the large-scale promotion of banking outsourcing business; customers do not need to distinguish between channels for inquiries, and requests submitted through any channel can receive a unified and standardized response; high-urgency inquiries are prioritized to avoid the backlog of customer requests; response templates are formulated based on the bank's standard business processes to ensure that the content of outsourced agents' replies is accurate and standardized, and to reduce customer complaints caused by inconsistent replies. Attached Figure Description
[0027] Figure 1 This is a schematic diagram of the overall structure of the present invention. Detailed Implementation
[0028] The present invention will now be further described in conjunction with specific embodiments. It should be noted that, without conflict, the various embodiments or technical features described below can be arbitrarily combined to form new embodiments.
[0029] In the description of this invention, it should be noted that directional terms such as "center," "lateral," "longitudinal," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise" indicate the orientation and positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. They should not be construed as limiting the specific protection scope of this invention.
[0030] It should be noted that the terms "first" and "second" in the specification and claims of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.
[0031] One preferred embodiment of the present invention, such as Figure 1 As shown, the intelligent triage and response system for multi-channel customer inquiries for bank outsourcing includes a multi-channel access module, an inquiry preprocessing module, a hierarchical triage module, an outsourcing response module, a rule management module, and an encrypted storage module. The modules interact with each other through an internal bus.
[0032] The multi-channel access module is used to simultaneously access consultation requests from telephone voice, online banking, mobile banking, WeChat official accounts and SMS channels, parse the protocol format of each channel and convert it into a unified data structure;
[0033] The consultation preprocessing module is connected to the multi-channel access module and performs noise filtering and keyword extraction on consultation requests. The keyword extraction is based on a preset financial business dictionary for precise matching.
[0034] The hierarchical triage module is connected to the consultation preprocessing module and the rule management module respectively. It calls the preset triage rule library to prioritize and assign seats to consultation requests. The triage rule library includes three-dimensional judgment logic of channel weight, business type and urgency.
[0035] The outsourced response module connects to the hierarchical distribution module, pushes consultation information and response templates with desensitized labels to the outsourced agent terminals, receives agent feedback and synchronizes it to the bank's core system;
[0036] The rule management module is used by bank administrators to configure traffic dictionaries, update business dictionaries and response templates, and supports setting permission isolation based on the outsourced team dimension.
[0037] The encrypted storage module performs symmetric encryption storage on consultation records, diversion logs, and anonymized customer data, and only grants complete data query permissions to authorized bank terminals.
[0038] The multi-channel access module includes a protocol adaptation sub-module, which presets parsing rules for more than 20 mainstream channel communication protocols and supports expanding adaptation capabilities through configuration files when adding new channels without modifying the core code.
[0039] The noise filtering process of the consultation preprocessing module is as follows: first, remove duplicate characters and meaningless auxiliary words; then, filter valid consultation content through a character length threshold; and finally, mark high-priority requests containing keywords such as complaint, emergency, and freeze.
[0040] The three-dimensional decision logic of the hierarchical flow division module is as follows:
[0041] Channel weighting: Telephone channel weight (1.2) > Mobile banking channel weight (1.0) > WeChat and online banking channel weight (0.8) > SMS channel weight (0.6);
[0042] Business type dimension: Match agent business tags in the order of account anomaly, credit consultation, wealth management business, and regular inquiry;
[0043] Urgency level: Three priority levels are set based on keyword matching results, with high priority requests having a response time limit of ≤15 seconds.
[0044] The outsourced response module includes a template calling submodule. The response template adopts a fixed text and variable placeholder structure. The variable placeholders are automatically filled with de-identified data obtained from the bank's core system. The de-identified data masks the customer's complete account number, ID number, and transaction amount details.
[0045] The process for obtaining the de-identified data is as follows: the outsourced response module sends a query request with a session identifier to the bank's core system. After verifying the validity of the session, the core system returns the de-identified data. The data transmission is encrypted using the TLS 1.3 protocol.
[0046] The rule management module includes a version control unit, which retains historical versions of modifications to traffic diversion rules and response templates, supports rollback to the three most recent valid configurations, and all modifications must be approved by two bank personnel before taking effect.
[0047] The hierarchical traffic distribution module also includes a load balancing unit, which collects the current consultation processing volume and historical response time of outsourced agents in real time. When the processing volume of a single agent exceeds a preset threshold, new requests are automatically distributed to low-load agents with the same business tag.
[0048] The outsourced response module also includes an exception handling unit. When a consultation request cannot match the keywords in the business dictionary, it automatically triggers a manual escalation process, transfers the request to the bank's internal customer service, and synchronizes the collected consultation context data.
[0049] The encrypted storage module adopts a timed backup mechanism, performing incremental backups every day at midnight. Backup data is retained for more than 90 days, and backup files can only be restored after verification by the bank's Ukey.
[0050] Working principle:
[0051] In use, the system first connects to commonly used customer inquiry channels such as telephone, online banking, mobile banking, WeChat official accounts, and SMS through a multi-channel access module. Addressing the differences in communication formats across different channels, the module has built-in preset protocol parsing rules that automatically convert various inquiry requests into a unified data format, avoiding information confusion caused by different channels and enabling a single entry point to collect all inquiries. The preprocessing module streamlines and refines the unified format inquiry requests, first automatically removing meaningless auxiliary words such as duplicate characters, then filtering valid content based on a length threshold of 5-500 characters, and finally precisely matching keywords against the bank's preset financial business dictionary. Simultaneously, high-priority requests containing words such as "complaint" and "urgent" are marked to prepare for subsequent traffic routing.
[0052] The tiered and traffic-distribution module calls the administrator's preset three-dimensional rule base to determine and allocate consultation requests layer by layer. The first step is to sort by channel weight; the second step is to match the business type and assign the consultation to the outsourced agent with the corresponding business tag; the third step is to mark the urgency based on keywords, requiring a response within 15 seconds for high-priority requests. Meanwhile, the module monitors agent load in real time. If a single agent processes more than 8 requests per hour, it automatically forwards new requests to agents with the same business tag and lower load. If a business tag cannot be matched, the request is directly transferred to the bank's internal customer service to avoid outsourced agents handling inquiries beyond their capacity. After receiving an assigned inquiry, the outsourced agent's terminal automatically calls the corresponding business's response template and requests anonymized customer data from the bank's core system. This data is then encrypted and transmitted to fill the template placeholders. Agents can supplement personalized responses based on the template, and the feedback data is synchronized to the bank's core system. Bank administrators can configure traffic distribution weights, update business dictionaries, and response templates at any time through the rules management module. All modifications require dual review and historical versions must be retained, with rollback support. The encrypted storage module performs symmetric encryption on inquiry records, traffic distribution logs, and other data. Only authorized bank terminals can query the complete data, while outsourced terminals can only view anonymized information. Data is automatically incrementally backed up every day at midnight and retained for 90 days for future reference.
[0053] The basic principles, main features, and advantages of this invention have been described above. Those skilled in the art should understand that this invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made without departing from the spirit and scope of the invention, and all such changes and modifications fall within the scope of the invention as claimed. The scope of protection claimed by this invention is defined by the appended claims and their equivalents.
Claims
1. A multi-channel customer inquiry intelligent triage and response system for bank outsourcing, characterized in that: It includes a multi-channel access module, a consultation preprocessing module, a hierarchical traffic distribution module, an outsourced response module, a rule management module, and an encrypted storage module. The modules interact with each other through an internal bus. The multi-channel access module is used to simultaneously access consultation requests from telephone voice, online banking, mobile banking, WeChat official accounts and SMS channels, parse the protocol format of each channel and convert it into a unified data structure; The consultation preprocessing module is connected to the multi-channel access module and performs noise filtering and keyword extraction on consultation requests. The keyword extraction is based on a preset financial business dictionary for precise matching. The hierarchical triage module is connected to the consultation preprocessing module and the rule management module respectively. It calls the preset triage rule library to prioritize and assign seats to consultation requests. The triage rule library includes three-dimensional judgment logic of channel weight, business type and urgency. The outsourced response module connects to the hierarchical distribution module, pushes consultation information and response templates with desensitized labels to the outsourced agent terminals, receives agent feedback and synchronizes it to the bank's core system; The rule management module is used by bank administrators to configure traffic dictionaries, update business dictionaries and response templates, and supports setting permission isolation based on the outsourced team dimension. The encrypted storage module performs symmetric encryption storage on consultation records, diversion logs, and anonymized customer data, and only grants complete data query permissions to authorized bank terminals.
2. The intelligent triage and response system for multi-channel customer inquiries for bank outsourcing as described in claim 1, characterized in that: The multi-channel access module includes a protocol adaptation sub-module, which presets parsing rules for more than 20 mainstream channel communication protocols and supports expanding adaptation capabilities through configuration files when adding new channels without modifying the core code.
3. The intelligent triage and response system for multi-channel customer inquiries for bank outsourcing as described in claim 1, characterized in that: The noise filtering process of the consultation preprocessing module is as follows: first, remove duplicate characters and meaningless auxiliary words; then, filter valid consultation content through a character length threshold; and finally, mark high-priority requests containing keywords such as complaint, emergency, and freeze.
4. The intelligent triage and response system for multi-channel customer inquiries for bank outsourcing as described in claim 1, characterized in that: The three-dimensional decision logic of the hierarchical flow division module is as follows: Channel weighting: Telephone channel weight (1.2) > Mobile banking channel weight (1.0) > WeChat and online banking channel weight (0.8) > SMS channel weight (0.6); Business type dimension: Match agent business tags in the order of account anomaly, credit consultation, wealth management business, and regular inquiry; Urgency level: Three priority levels are set based on keyword matching results, with high priority requests having a response time limit of ≤15 seconds.
5. The intelligent triage and response system for multi-channel customer inquiries for bank outsourcing as described in claim 1, characterized in that: The outsourced response module includes a template calling submodule. The response template adopts a fixed text and variable placeholder structure. The variable placeholders are automatically filled with de-identified data obtained from the bank's core system. The de-identified data masks the customer's complete account number, ID number, and transaction amount details.
6. The intelligent triage and response system for multi-channel customer inquiries for bank outsourcing as described in claim 1, characterized in that: The process for obtaining the de-identified data is as follows: the outsourced response module sends a query request with a session identifier to the bank's core system. After verifying the validity of the session, the core system returns the de-identified data. The data transmission is encrypted using the TLS 1.3 protocol.
7. The intelligent triage and response system for multi-channel customer inquiries for bank outsourcing as described in claim 1, characterized in that: The rule management module includes a version control unit, which retains historical versions of modifications to traffic diversion rules and response templates, supports rollback to the three most recent valid configurations, and all modifications must be approved by two bank personnel before taking effect.
8. The intelligent triage and response system for multi-channel customer inquiries for bank outsourcing as described in claim 1, characterized in that: The hierarchical traffic distribution module also includes a load balancing unit, which collects the current consultation processing volume and historical response time of outsourced agents in real time. When the processing volume of a single agent exceeds a preset threshold, new requests are automatically distributed to low-load agents with the same business tag.
9. The intelligent triage and response system for multi-channel customer inquiries for bank outsourcing as described in claim 1, characterized in that: The outsourced response module also includes an exception handling unit. When a consultation request cannot match the keywords in the business dictionary, it automatically triggers a manual escalation process, transfers the request to the bank's internal customer service, and synchronizes the collected consultation context data.
10. The intelligent triage and response system for multi-channel customer inquiries for bank outsourcing as described in claim 1, characterized in that: The encrypted storage module adopts a timed backup mechanism, performing incremental backups every day at midnight. Backup data is retained for more than 90 days, and backup files can only be restored after verification by the bank's Ukey.