To address the challenges of low precision in traditional hotel marketing and insufficient customer value extraction within the hospitality industry

this paper establishes a big data-driven hotel customer profiling system along with a precise marketing implementation plan. By integrating various internal business datasets from hotels

this study develops a tagging system comprising four dimensions; furthermore

by incorporating the length of stay as an additional factor on top of the existing RFM model

a new RFMD customer value evaluation model is constructed. The customer segmentation process employs the Analytic Hierarchy Process (AHP) and the K-means clustering algorithm. An empirical validation was conducted using 15

246 active customer records from a specific hotel chain spanning the period from 2023 to 2024. The results demonstrate that customers can be effectively segmented into five distinct groups

and targeted marketing campaigns based on these customer profiles achieve a conversion rate 5.1% higher than that of traditional models

while improving the return on investment (ROI) by 32.7%. This approach provides hotels with a reusable technical pathway and a solid decision-making basis for their digital marketing transformation.