Holy Stoneoly Stone was founded in 2009, focusing on drones, baby products, and children's toys in the e-commerce sector. The company has offices in Shenzhen, Fujian, and Xiamen, and it excels in cross-border e-commerce, particularly in the development and sales of quadcopter drones. Holy Stone is the top brand in the drone category on Amazon.
With the motto "small city, big dreams," Holy Stone aims to become a global enterprise and a company that provides its employees with the highest happiness.
Holy stone Technology x VOC AI: Seamless Cross-Time Zone Customer Service with AI
As a leading brand in toy drones, Holy Stone's customer service team is responsible for serving a vast consumer base across different time zones and regions. With the expansion of product categories and business growth, both domestically and internationally, the pressure on customer service has increased significantly. For a company committed to providing an excellent shopping experience and quality service, timely, error-free, and customer-focused service is crucial. To address these challenges, VOC AI collaborated with Holy Stone's customer service department, enabling a dedicated AI customer service system to serve Holy Stone customers effectively.
Consumer Interaction Scenarios: AI Intent Identification Personalized Solutions for Daily Operational Tasks, Fast and Accurate
Historically, responding to customer tickets required significant time and effort. For instance, during new operational activities, the customer service team would spend a considerable amount of time answering similar questions. For example, when operational activities like coupon distribution, social media interactions, or discounts are launched, traditional smart customer service systems might handle some direct responses but would struggle with more complex consumer interactions. If a consumer asks about a product issue but receives a response about coupon distribution, it can lead to customer dissatisfaction.
VOC AI, leveraging advanced models, prioritizes identifying the intent behind customer messages. It categorizes and responds to customer inquiries based on their specific needs, generating human-like and comfortable responses. Even when consumers provide partial information, the AI avoids repeating questions, thus handling these scenarios more effectively and friendlier.
Holy Stone's Senior Customer Service Advisor noted: " Previously, responding to these tickets would take 3.5 hours; now it only takes half an hour for checking. The AI's responses to consumers are generally correct. For situations that the AI cannot handle, it informs human customer service for timely intervention, freeing up our hands and reducing stress significantly."
Fault Diagnosis Scenarios: Enhanced Fault Diagnosis with Latest AI Models
Timely and Accurate Responses, Preparing for Peak Seasons
Drones have diverse and complex fault issues, ranging from components, GPS, drones themselves, batteries, to APP connections and camera issues. During the GPT-3.5 era, even with extensive knowledge bases, handling complex fault diagnosis was challenging. For example, when consumers describe issues like "the battery is fine, but other problems persist," earlier models struggled to respond adequately.
VOC AI used the latest GPT-4 model, combined with deep understanding and rich practical experience, to fine-tune the AI. This approach minimized repetitive inquiries based on the information provided by consumers and answered remaining questions using the knowledge base.
Holy Stone's Customer Service Manager stated: "Thanks to VOC AI taking over customer interaction scenarios, we can focus our manpower on knowledge base management. Currently, AI can automatically handle preliminary fault diagnosis. As peak seasons like Black Friday and Christmas approach, fault feedback will increase exponentially. Timely responses to customer needs will positively impact return rates. Preparing and training AI in advance will help us better handle these escalating customer complaints."
Purchase Consultation and Negative Review Tracking: Human and AI Collaboration for Enhanced Customer Experience
In every e-commerce platform, consumers have the right to express their opinions and feedback, whether positive or negative. Each product has its strengths and areas for improvement, leading consumers to leave reviews based on their experiences. On Amazon, negative reviews are automatically transferred to the seller, and Holy Stone is no exception. If these negative reviews are not handled promptly, they can directly impact product sales.
In this scenario, VOC AI designed a combined AI and human service approach for Holy Stone. This flexible handling scheme satisfies consumers' actual needs, avoiding situations where AI over-promises and increases the cost of handling negative reviews.
Human customer service, with professional and flexible handling schemes, timely addresses and resolves negative review issues. Meanwhile, AI acts as a continuous follow-up and customer service manager, automating the follow-up process until consumers accept the solution and respond to emails. Even when human customer service is busy, every consumer's grievance is addressed with AI support.
Customer Service Beyond After-Sales Customer inquiries are not limited to after-sales issues. A significant portion involves pre-purchase product consultations. VOC AI provides pre-purchase solutions by identifying the products or accessories consumers wish to buy and automating the process to retrieve stock and purchase links, quickly feeding back to consumers. This ensures 24/7 protection of business opportunities.
The introduction of the AI customer service system has significantly improved Holy Stone's work efficiency and customer experience, making complex service scenarios more streamlined and accurate. Human resources are now more focused on knowledge upgrade and customer experience enhancement. In the future, the collaboration between VOC AI and Holy Stone will continue to explore deeper customer experience scenarios, leveraging AI to support Holy Stone's global development.
