DESCRIPTION
The project was carried out at a travel agency operating with nationwide coverage, offering a wide portfolio of various travel packages, accommodation bookings, and car rental services to its clients. The company received hundreds of inquiries daily over the phone, a significant portion of which were related to questions about car rentals. The customer service team worked under constant pressure, and incoming calls often aimed to clarify repetitive standard information: rental conditions, insurance options, deposits, car categories, opening hours, pick-up points, modification options. DetSystems aimed to create a voice-based AI chatbot that could fully handle these inquiries without the involvement of a human operator. The result was an intelligent, naturally sounding system that provides accurate, quick, and consistent answers to all incoming car rental inquiries based on the agency's own knowledge base and detailed internal information.
PROBLEM
One of the biggest difficulties faced by the travel agency was that the phone customer service traffic was constantly high, and the workload was very unevenly distributed throughout the day. Most incoming calls did not require expert intervention, yet they took up a lot of time from the agents, which meant they could not dedicate enough time to more complex, higher-value cases. The system was often overloaded, and the information related to car rentals was frequently repeated, resulting in additional unnecessary work hours.
The problem was further exacerbated by the fact that the services of the office contained different rules by country, partner company, and package, making it a time-consuming task for employees to provide accurate information. It became clear to management that the current operation was not scalable in the long run, and the system was unable to handle the increasing demand without significantly increasing the customer service staff. Car rental was one of the most sought-after services in the portfolio, so it was clear that automation was needed in this area as the first step. There was a need for a solution capable of communicating in a human-like manner, interpreting questions, providing relevant answers, and doing so with a knowledge base that was always up-to-date and accurate.
SOLUTION
DetSystems has developed a voice-based AI agent capable of independently receiving incoming calls, interpreting the questions posed by users, and providing accurate and personalised responses based on the detailed and customised knowledge base of the office. The system did not simply follow a basic IVR logic but employed natural speech recognition technologies, allowing it to handle freely formulated questions and intelligently follow the flow of conversation. During development, we built the knowledge base in layers: from general information to country-specific rental terms and the unique rules of partner companies, all relevant details were incorporated to provide comprehensive customer information regarding car rentals.
The system was able to handle the most common situations: users could inquire about car categories, pricing, insurance options, pickup times, modification possibilities, or even special requests. The agent always relied on predetermined internal information, ensuring that the responses were not only quick but also consistently reflected the office's operations. One of the most significant achievements of the development was that the system could handle real phone calls in a natural voice, meaning that for customers, the service was indistinguishable from a conversation with a human operator.
The solution was also capable of continuously learning and evolving based on user interactions. After every conversation, relevant information was fed back into the system, allowing us to fine-tune the responses and speech recognition logic. As a result, the AI agent, just weeks after deployment, was noticeably faster, more accurate, and provided smoother conversations than the initial version.
RESULT
The introduction of a voice-based AI chatbot has radically transformed the customer service operations of the travel agency. A significant portion of calls has become manageable automatically, relieving the customer service team and allowing them to focus on more complex, higher-value issues instead of previously monotonous, repetitive tasks. Callers received instant answers to all their car rental-related questions, without waiting, even during nighttime hours. The error rate has been minimised, as the system always worked precisely from the provided knowledge base, eliminating misunderstandings or misinformation.
Operations have become scalable: the system is capable of handling any number of incoming calls simultaneously, meaning that seasonal peak periods no longer cause congestion or increased workforce demands. For management, it was an outstanding achievement that the agent not only became a digitalised customer service element but also an intelligent system that provided sustainable long-term operation, predictable quality, and significant cost savings.
The project demonstrated the power of voice-based AI agents when built on a company's internal knowledge: customer communication became faster, more accurate, and significantly more efficient, while completely new innovation gateways opened up for the agency.
