RHINOSHIELD Making AI the Second Brain of Customer Support

Since the debut of OpenAI’s ChatGPT at the end of 2022, an AI revolution has officially swept the globe. Major enterprises have begun to reflect: Exactly what substantial changes can AI bring to brand operations?

RHINOSHIELD, a Taiwanese mobile accessory brand focused on material science, took the lead during this wave of innovation. Aiming to solve customer service challenges arising from peak/off-peak seasons and cross-regional operations, they partnered with Raccoon AI to integrate artificial intelligence into their daily operations. This collaboration has progressively realized the structuring of internal knowledge and a comprehensive transformation of the external customer experience. While reducing the labor burden on the support team, they have achieved more efficient and stable service quality, reaching an 80% AI satisfaction rate.

The Raccoon AI team specifically invited Ku Chien-liang (古健良), Project Manager at RHINOSHIELD, to share the brand’s journey during the early stages of AI implementation. He discusses how adopting Raccoon AI effectively boosted customer service capacity and overall operational efficiency. Through their experience, we can identify the key elements brands need when driving AI transformation. Let’s explore their insights to find new inspiration for AI-driven growth.

Thinking of "Assistance" rather than "Replacement": RHINOSHIELD Sees a Turning Point for Customer Service Efficiency within AI Trends.

Actually, in the early stages of implementing AI, our team's original intention wasn't to have it simply reply to questions automatically," said Ku Chien-liang. "Instead, we hoped that through AI, we could truly help our Customer Service (CS) agents find solutions more quickly, improve efficiency, and even address the needs of customers in different languages within a cross-regional context.

Reflecting on the Reason for the decision to implement AI, it can be traced back to the end of 2022 when ChatGPT shocked the world. At that time, major international enterprises began exploring and considering how AI could bring more diverse transformations to brand operations.

RHINOSHIELD also saw an opportunity to boost customer service efficiency: Could AI be used to handle some of the repetitive tasks, assisting the customer service team in responding to customers faster and with greater precision?

To ensure the AI implementation was truly effective, the RHINOSHIELD team first set out to identify the current challenges facing their customer service department. "The mobile accessory industry experiences very distinct peak and off-peak seasons, with a large portion of our sales concentrated in the fourth quarter,"

Ku Chien-liang explained. "Consequently, the surge in demand during the year-end peak season creates a massive backlog of inquiries, leading to a high demand for labor. However, maintaining that same level of staffing during the off-peak season would create a burden in terms of personnel costs.

Therefore, our primary goal for AI was to figure out how to properly allocate customer service resources across different seasons and boost efficiency. We expect AI to assist with basic inquiries so that the team can maintain a stable quality of service, even during the peak season.

RHINOSHIELD's business footprint spans the globe. The brand currently operates 11 official websites worldwide, covering more than 8 major international languages. In the past, when customer service agents encountered inquiries from customers in different languages, they often had to rely on the assistance of native speakers from various regions. However, this also invisibly increased communication costs.

By effectively leveraging AI's linguistic capabilities, the brand can more efficiently achieve cross-regional customer service integration, ensuring that customers in different markets receive immediate and accurate service experiences. In light of these combined challenges and opportunities, RHINOSHIELD became even more determined to implement AI, accelerating the pace of the brand's digital transformation in customer service.

However, at that time, various AI solutions were springing up in the market like mushrooms after a rain. Consequently, the team began by researching their existing partner, Zendesk, while simultaneously evaluating multiple other systems available. Ultimately, they decided to partner with Raccoon AI, aiming to integrate its capabilities into their existing Zendesk framework. By leveraging Raccoon AI's products and technology, they sought to enhance overall customer service efficiency and provide a more seamless experience for their customers.

Since our founder places great importance on AI applications, the first project we collaborated on with Raccoon AI was an 'Internal AI' application," Ku Chien-liang shared.

RHINOSHIELD 官網

Identifying the "Key Person" to drive implementation is the starting point for a brand’s journey toward "structuring" and "digitizing" its knowledge.

During the implementation process, we realized that the greatest challenge was determining how to organize the SOPs and documents needed to train the AI," said Ku Chien-liang.

In the early stages of implementation, the RHINOSHIELD team’s expectations for AI technology were no different from those of most enterprises. They initially assumed that the implementation process would be similar to using everyday generative AI tools—simply providing a link to the brand’s website, allowing the AI to automatically crawl and learn the data, and then immediately deploying it for practical use.

However, once execution began, they discovered it was not as simple as imagined. "We originally thought that as soon as the technical team completed the integration, the customer service team would instantly see a boost in efficiency," Ku Chien-liang recalled. "But we quickly realized that much of the critical information actually resided within the personal experience of the staff. For instance, how to query orders in the internal system or how to handle specific service workflows—these things were traditionally stored in 'human memory,' and we lacked the clear SOPs and documented records necessary for the AI to learn from.

Transforming the brand’s knowledge through 'Structuring' and 'Digitization' was the turning point in RHINOSHIELD’s AI journey, allowing the technology to reach its full potential.

The second challenge: Who should be responsible for writing these SOPs? And whose standards should serve as the primary reference?

Ku Chien-liang explained, our customer service agents focused primarily on solving the problem as their top priority. Asking them to find extra time during their busy schedules to translate the processes in their heads into SOP documents was quite difficult to push internally.

Beyond the issue of employee motivation, the customer service process itself is extremely complex. The logic of a customer interaction is actually like a branching mind map. For example, when a customer asks 'Question A,' the agent must determine the next step based on the customer’s response and provide a corresponding answer. Every reply path requires further judgment, which constantly branches out into more sub-questions. This significantly increased the complexity of gathering information and standardizing processes—not to mention the multilingual and multi-national knowledge we had to manage. Each region's operational flow varies based on local conditions, making it honestly quite difficult to solve completely.

In the past," Ku Chien-liang explained, "our customer service agents focused primarily on 'solving the problem' as their top priority. Asking them to find extra time during their busy schedules to translate the processes in their heads into SOP documents was quite difficult to push internally.

Beyond the issue of employee motivation, the customer service process itself is extremely complex. The logic of a customer interaction is actually like a branching mind map. For example, when a customer asks 'Question A,' the agent must determine the next step based on the customer’s response and provide a corresponding answer. Every reply path requires further judgment, which constantly branches out into more sub-questions. This significantly increased the complexity of gathering information and standardizing processes—not to mention the multilingual and multi-national knowledge we had to manage. Each region's operational flow varies based on local conditions, making it honestly quite difficult to solve completely.

Facing the challenges of staff mindsets and the complex translation of multi-layered logic proved to be a daunting hurdle for RHINOSHIELD in their push for AI. However, Ku Chien-liang shared the brand's experience in breaking this deadlock: "The most important thing is to first identify a 'Key Person' who is genuinely willing to drive the initiative forward!

Based on RHINOSHIELD’s experience, customer service staff who are highly receptive to emerging technologies and not confined by traditional operational frameworks are ideal for this role. Such a person can not only gain a deep understanding of customer service workflows but is also willing to embrace the changes brought about by new technology. Therefore, identifying this candidate and fully empowering them with sufficient authority and resources is a vital part of an enterprise's push for AI implementation.

After finding the Key Person and providing them with sufficient empowerment and resources, the next critical step is 'Rapid Iterative Testing,'" Ku Chien-liang shared. "Once we have handed over the knowledge data for the AI to learn, the most important thing is to immediately verify whether it can effectively solve customer problems. At this stage, Raccoon AI’s external Widget was a huge help. After uploading our knowledge documents, we could use the chatbox in the bottom-left corner of the website to simulate customer inquiries and check if the AI's initial responses were correct. This real-time testing mechanism allowed us to quickly identify gaps in responses, continuously optimize our SOPs, and significantly enhance both customer service quality and training efficiency.

Implementing Raccoon AI Resolves Over 85% of Basic Inquiries, Driving a Tangible Boost in Customer Service Team Efficiency and Quality

From reviewing the individual conversation logs in the Raccoon AI backend, it’s clear that a significant portion of our basic inquiries are being resolved. This has truly helped us achieve our original objectives," Ku Chien-liang stated with confidence.

I believe Raccoon AI’s model of offering conversational AI as a subscription-based, ready-to-use service is excellent. The platform is remarkably simple and intuitive; for instance, I was able to get up to speed quickly during my first use without any formal training. Furthermore, as long as you have a basic grasp of system logic, you can easily train the AI yourself

Furthermore, when discussing system upgrades, localized support, and technical collaboration, Ku Chien-liang specifically commended the professionalism and responsiveness of the Raccoon AI team. "Whenever we propose new requirements or identify areas for improvement, Raccoon AI’s technical team is able to quickly adjust features and resolve issues, helping us achieve our goals faster," he stated.

He further pointed out that Raccoon AI has been deeply rooted in the Taiwan market for years, offering a high degree of local integration. Beyond seamless connectivity with major customer service systems like Zendesk, they also provide robust integration with platforms commonly used in Taiwan, such as Omnichat and LINE. "Looking at consumer behavior in the Taiwan market, Raccoon AI’s services cover over 90% of the most frequently used customer service channels. This brings substantial, practical benefits to our support team," Ku said.

Raccoon AI recently launched "Text to Flow," an AI-powered intelligent script editor. This tool allows users without a programming background to create brand-specific scenarios and dialogues simply by dragging and dropping settings. It also offers advanced AI conversation scenario configurations. We look forward to this tool providing the RHINOSHIELD team with even deeper, more specialized scenario applications.

Driving AI Workflows from the 'Outside In': Establishing a Comprehensive Knowledge Base is the Stepping Stone to Successful Enterprise AI Implementation

If I were to go through this AI implementation process again, I would recommend starting with customer-facing AI support," Ku Chien-liang stated with confidence. "This is because it directly impacts the efficiency of the customer service department and is relatively easier to push forward during the initial stages of implementation.

Because conversational AI tools can automatically resolve basic consumer inquiries— such as 'How do I track my order?' or 'How do I cancel or modify my order?'—they serve as the first step in boosting efficiency. For RHINOSHIELD, even though comprehensive tutorials for these tasks exist on our website, we still receive a massive volume of these redundant questions during peak sales seasons. By utilizing tools like Raccoon AI’s widget and simply uploading existing FAQ data, we can prioritize and resolve these low-difficulty inquiries first

The next phase involves establishing the internal corporate knowledge base," Ku Chien-liang noted. "By starting with customer-facing AI, we can observe which team members demonstrate the most initiative and insight regarding AI implementation. These individuals are potential candidates to be nurtured into the Key Person for internal AI rollout. Once they are empowered with the necessary authority and resources, I believe the pace of organizational change will accelerate significantly

Ku Chien-liang further added that, in his view, company size has little correlation with the difficulty of implementing AI; rather, it boils down to the proactive establishment of a "Corporate Knowledge Base.

Data such as customer service SOPs or knowledge should be documented as early as possible," he said. "While a company might grant staff some flexibility when handling customer issues, there is always an underlying SOP. By digitizing and structuring these SOPs as much as possible, you build the necessary infrastructure for AI implementation. When you have rich data available for training the AI, it inevitably accelerates the time it takes to apply AI to actual work and makes internal promotion smoother. I believe that the 'digitization and structuring' of the corporate knowledge base is the most common pain point enterprises face when adopting AI.

I believe the way customer service professionals work will undergo a massive transformation in the future. Their role may no longer involve responding to customers directly—or perhaps only a small minority will do so. Instead, the team’s workflow will pivot toward 'AI Optimization' as the core premise. They will focus on continuously enriching the AI’s knowledge base and SOPs while simultaneously analyzing customer inquiries to identify opportunities for improvement. This shift is essential to capturing more business opportunities.

Through this in-depth interview with RHINOSHIELD, it is clear that their starting point for AI implementation was never just about introducing a new system for the support team. Instead, it was about redefining how an enterprise operationalizes AI through a shift in mindset. From initially auditing the pain points of customer service workflows to successfully deploying cross-regional and multi-lingual applications, this journey represents far more than just process innovation; it is a profound transformation rooted in the "structuring" and "digitization" of organizational knowledge.

RHINOSHIELD has demonstrated through concrete action that implementing AI is not about replacing customer service staff, but about transforming technology into the team's most powerful support system. Only by having the courage to take that first step toward embracing AI can an enterprise find a new equilibrium between efficiency, agility, and service quality—ultimately unlocking the next milestone of operational growth.

After reviewing this case study, if you are interested in Raccoon AI’s services and would like to explore how AI can alleviate the pressure on your customer service team, you are welcome to start a trial or consult with our professional advisors to learn more.

Schedule a consultation now

---