<p>Benchmarking in service industries has received considerable attention; however, traditional approaches predominantly rely on customer surveys and financial or operational metrics. These methods are often time-consuming, resource-intensive, and limited in capturing unexpected yet impactful service attributes derived from actual customer experience. To address these limitations, this study proposes a novel customer-centric, data-driven framework that transforms the traditionally qualitative and manual process of competitor definition, performance diagnosis, and priority setting into an objective, systematic pipeline using large-scale online customer reviews. The framework consists of four key components: (1) topic modeling to identify service attributes from unstructured review texts, (2) index and sentiment analysis to assess the importance of each attribute and evaluate performance, (3) k-means clustering and TOPSIS to identify relevant competitors and best practices, and (4) importance-performance competitor analysis (IPCA) to develop targeted strategic actions. A case study using 26,934 reviews from 26 hotels in Bangkok demonstrates the practical utility and scalability of the proposed framework. This research contributes to marketing analytics by offering a systematic, customer-perception-driven alternative to traditional benchmarking, supporting continuous service improvement and competitive positioning in dynamic digital markets, as demonstrated through its application to Bangkok’s hotel industry.</p>