Transforming customer opinions into actionable insights used to take a lot of time. Hong Kong Express (HK Express) announced that it has used Google Cloud’s Vertex AI and BigQuery services to significantly improve operational efficiency through advanced technologies such as artificial intelligence big data analysis, especially in the processing and analysis of customer opinions, with efficiency increased by 41%.
Hong Kong Express pointed out that it continues to improve passengers’ flight experience and the company actively collects and analyzes customer opinions. However, with the expansion of service scope and the explosive growth of tourism demand, traditional data processing methods can no longer meet the rapidly changing market demand. Google Cloud’s advanced data analysis and artificial intelligence platform has become an assistant for Hong Kong express transportation. Chen Yingqian, Director of Digital and Information Technology at Hong Kong Express, said: “By collaborating with Google Cloud, we have successfully improved the efficiency of our data science team, streamlined the process of handling customer feedback, and also brought rich insights to the company to drive business development. , enabling us to provide excellent services that better meet our customers’ needs.”
HK Express’s Journey Experience & Design (JED) department is responsible for processing a large number of passenger survey results, including up to 10,000 passenger opinion surveys per month. Since each response may have both positive and negative opinions, manual screening and classification have been relied upon in the past.Luk Wai-yin, Head of Digital Transformation and Insights at Hong Kong Express Airlines, said: “For example, passengers may praise the cabin crew for their excellent service, but at the same time point out that there is room for improvement in the temperature of the in-flight meals. Such opinions need to be included in two different Categories: One is about crew attitude, and the other is about in-flight dining experience. After completing the first round of classification, a second round of analysis is needed to confirm which food on the menu these opinions are about, in order to narrow down the food experience. analysis. This manual process is not only tedious and time-consuming, but also prone to human error.”
In the past, manual screening was used
The JED department originally had to manually screen passengers’ opinions one by one and classify the responses as positive or negative.Hong Kong ExpressUse Google Cloud’s BigQuery service to build a unified big data analysis platform andContinuously fine-tune the topics, subtopics list and instructions for AI models on BigQuery every week to ensure that the system continues to improve over time and is comprehensiveIt speeds up data processing and also improves the accuracy of data processing. Through the application of Vertex AI, the process is fully automated. Vertex AI can not only accurately identify and classify customer opinions, but also segment them based on the specific content of the opinions, such as food quality or service attitude, which greatly improves processing efficiency and accuracy. In addition, Vertex AI’s multi-language capabilities are particularly important for international airlines. It understands customer opinions in different languages and turns them into useful insights, helping companies better understand the needs and expectations of international customers.
It is reported that using Vertex AI to improve customer opinion surveys is part of Hong Kong Express’ blueprint to promote data innovation and customer service strategies. The company is also exploring the use of generative AI (GenAI) and Google’s latest large-scale language model to analyze the customer service platform The dialogue between the customer service director and the customer hopes to improve customer satisfaction and speed up problem resolution.
Source: Google Cloud
Tags: Hong Kong Express cooperates Google Cloud analyze customer opinions improve satisfaction ezone .hk Times Strategy
