To further compare the regional precipitation monitoring performance of radar network-based Quantitative Precipitation Estimation (QPE) products (from CAPPI data at different heights) across Guangdong Province, this study validated and compared hourly radar QPE products during Guangdong’s pre-flood season . The analysis utilized hourly gauge-observed precipitation data and employed evaluation metrics including Taylor statistics and comprehensive scores, with additional comparisons of cases across different precipitation types. The main conclusions were as follows: dual-polarization QPE outperformed single-polarization QPE across Guangdong Province, with its estimation accuracy improving as precipitation intensity increased; among all products, the dual-polarization QPE derived from 2.5 km-height CAPPI data (QPE_2.5) exhibited the best overall performance. QPE_2.5 and QPE_1.5 performed better in coastal and transitional regions, while QPE_2.5 and QPE_3.0 showed superior performance in inland regions. Case comparisons across precipitation types revealed that both single- and dual-polarization QPE achieved higher comprehensive scores for monsoon rainfall than for frontal rainfall. Specifically, dual-polarization QPE demonstrated more stable performance for monsoon rainfall—with QPE_2.0 being the most accurate for this type—whereas QPE_1.5 performed best for frontal rainfall.