The dependability of hydrological model simulations to produce reliable outcomes is influenced greatly by the accuracy and robustness of input data, particularly rainfall forcing estimates. In this regard, significant advancements in producing accurate rainfall products have been made in recent decades, ranging from rain-gauge-based measurements to remote-sensing-based estimates. One reliable product is the Climate Hazards Infrared Precipitation with Stations (CHIRPS), which utilizes infrared observations and incorporates in-situ station data. In this study, we evaluate the accuracy and robustness of CHIRPS products relative to the rain-gauge observations in the Bengawan Solo catchment, Indonesia. We compare the CHIRPS data for each available grid to the corresponding rain-gauge measurements within the same grid. The results indicate superior Pearson correlation coefficient (r) for CHIRPS grids that is covered by more than one rainfall station, as compared to those with only one rainfall station. On average, these grids with multiple rainfall measurements exhibit 36.5%, 11.2%, and 9.9% higher Pearson correlation coefficients for daily, monthly, and yearly time scales, respectively, compared to those with only one rainfall station within a grid. The analysis also reveals that the highest agreement is observed in monthly time-scale products, with an average r of 0.796, which is substantially higher than those of the daily (0.283) and yearly (0.673) products. These results highlight the significance of ‘rainfall-gauging’ various basins in developing countries even in already gauged basins, and is particularly even more critical in data-scarce areas, while also contribute to understanding the accuracy and robustness of CHIRPS data across different time scales.