Consumer finance, as a typical scenario with a large user base and complex business operations, urgently needs to leverage the intelligent capabilities of large models to improve service efficiency, optimize user experience, and strengthen risk control.
Customer Service Challenges: Traditional Robots "Can't Understand or Perform Well"
A well-known consumer finance company acted promptly and urgently needed to introduce a new generation of customer service intelligent agents to address the pain points of traditional customer service robots:
- Inaccurate intent recognition: It is unable to identify "I want to repay early" and "I want to repay ahead" as the same intent, and its comprehension ability is limited.
- Stagnant interaction experience: It can only mechanically answer fixed QA questions and cannot naturally switch to human assistance in complex situations.
- Limited format support: It does not support rich text, cards, and hyperlink navigation, and the path for customers to obtain information is lengthy and inefficient.
These problems are particularly prominent during business peak periods: On the core repayment day, the number of manual call sessions reaches 800 per day, and on ordinary days, there are 400 per day. However, there are only 9 fixed seats, and the human workload is extremely heavy, making it difficult to guarantee both efficiency and experience.
Solution: A large-scale model customer service system with a deeper understanding of consumer finance
To address the aforementioned pain points, this consumer finance company collaborated with Cloopen Ai to complete a comprehensive upgrade from the traditional customer service system to a large-scale model intelligent customer service system. They introduced agent Virtual Agent and an intelligent knowledge base. This is not merely a technical replacement; rather, it integrates the general capabilities of the large model with the business characteristics of consumer finance, achieving a truly intelligent service experience that "understands both the users and the business".
01Financial Inquiry: Asking More Gets Instant Understanding, Precise Responses
Current Issues: In the frequent financial inquiry scenarios of consumer finance, users' inquiries often have characteristics of "oralization and vagueness", such as: "When will the bill be released this month", "How much of the loan have I not repaid yet", etc. Traditional customer service either transfers to human agents due to inaccurate intent recognition, or can only return standardized responses, unable to solve the actual problems.
Solution: Based on the semantic understanding ability of large models, the Cloopen Ai agent Virtual Agent can deeply understand the diversity of customer needs, flexibly identify users' diverse expressions, and provide bill, repayment, overdue, etc. information in real time and accurately.
For example, no matter if the customer asks "pre-arranged repayment" or "repay earlier", the Virtual Agent can quickly understand their intentions and accurately provide repayment amount, process and channels, etc., achieving "understandable and accurate answers", and has successfully taken on 85% of daily inquiries at this consumer finance company, with customer satisfaction increasing by nearly 30%.
02Complex Consultation: Intelligent and flexible like a real person
Current Issues: When there are issues such as failed withdrawals, failed transfers, or sensitive account operations, traditional robots tend to give fixed template responses, lacking personalization and resulting in poor customer experience.
Solution: The Cloopen Ai agent Virtual Agent has capabilities such as large model context understanding and information correction based on semantic understanding. It covers various operations of loan accounts, such as failed withdrawals and failed transfers, and can provide personalized and heartwarming responses based on the relevant reasons returned by the interface and combined with fixed scripts.
In addition, for sensitive operations such as account unbinding, the Cloopen Ai Virtual Agent can also provide retention scripts and seamlessly provide operation entrances or transfer to human agents when the customer insists, ensuring a smooth and secure service experience.
03Post-loan management: Significantly alleviates peak human workload
Current problem: The original traditional robot system of this consumer finance company was unable to handle complex requests such as "I want to negotiate repayment", and relied on human customer service for judgment and handling.
Solution: The Cloopen Ai agent Virtual Agent deeply integrates the financial knowledge base and business logic. In the post-loan management scenario, it can intelligently identify requests like "I want to negotiate repayment" from users and automatically call the write-off status query interface to achieve efficient response.
If there is a written-off loan, it triggers a transfer to human service, providing personalized negotiation plans; if there is no written-off loan, it directly returns the question-and-answer content, guiding users to understand repayment options, achieving precise diversion and significantly alleviating the peak human workload.
04Account operation: Card-based guidance, smooth experience
Current problem: The original traditional customer service robots of the financial service company were unable to provide real-time guidance when changing repayment accounts. They could only rigidly guide to manual assistance.
Solution: Cloopen Ai agent assistant Virtual Agent can provide rich text and card-based operation guidance when customers need it. It can even redirect to the APP/mini program page, and customers can complete the process by simply clicking. The experience is more intuitive and the path is shorter. This has increased the efficiency of customer service operations for the financial service company by 30%.
05Knowledge base operation: The more used, the smarter
Current problems: The customer service representatives are occupied with a large amount of repetitive inquiries, the knowledge base update is lagging behind, and the service iteration is slow.
Solution: By fully integrating with the customer service reporting system, the client manager APP, and the large model base, the Cloopen Ai agent Virtual Agent can continuously learn and optimize based on customer feedback and conversation data. It can comprehensively analyze the knowledge gaps in the service process and promptly identify which contents need to be supplemented and improved, forming a virtuous cycle of continuous iteration and upgrading.
Currently, the Cloopen Ai agent has successfully covered 85% of the daily consultation questions of this consumer finance company, reducing the rate of transfer to human agents by 50%, allowing the human agents to be liberated from repetitive tasks and focus on handling more complex scenarios of higher value.
06Zero-loss switching: Ensuring business continuity
In this consumer finance project, Cloopen Ai achieved a complete migration of historical conversation data and QA scripts from the original customer service system to the new large-model intelligent customer service system. Over a thousand historical question-answer data and the three-level classification structure were all fully retained and adapted to the new system, with no data loss at all.
The entire migration process achieved a seamless business switch without any impact, and the system operated stably and reliably, fully guaranteeing service continuity.
With rich experience in the financial industry, Cloopen Ai has successfully helped many banks, consumer finance companies, and other financial companies complete the smooth migration from small models to large-model systems.
Dual improvement in experience and efficiency
The large-scale model intelligent customer service solution created by Cloopen Ai for this well-known consumer finance company not only enhances service quality and efficiency, but more importantly, through the deep understanding and integration of AI technology into business scenarios, it provides customers with a more intelligent, convenient, and secure financial service experience.
In the future, Cloopen Ai will continue to leverage the integration advantages of large-scale models and industry experience to assist more financial institutions in achieving high-quality and intelligent development.