The Equotip surface hardness tester is becoming a popular method for rock and stone weathering research. In order to improve the reliability of Equotip for on-site application this study tested four porous limestones under laboratory conditions. The range of stone porosity was chosen to represent likely porosities found in weathered limestones in the field. We consider several key issues: (i) its suitability for soft and porous stones; (ii) the type of probe required for specific on-site applications; (iii) appropriate (non-parametrical) statistical methods for Equotip data; (iv) sufficient sampling size. This study shows that the Equotip is suitable for soft and porous rock and stone. From the two tested probes the DL probe has some advantages over the D probe as it correlates slightly better with open porosity and allows for more controlled sampling in recessed areas and rough or curved areas. We show that appropriate sampling sizes and robust non-parametric methods for subsequent data evaluation can produce meaningful measures of rock surface hardness derived from the Equotip. The novel Hybrid dynamic hardness, a combination of two measuring procedures [single impact method (SIM) and repeated impact method (RIM)], has been adapted and is based on median values to provide a more robust data evaluation. For the tested stones in this study we propose a sample size of 45 readings (for a confidence level of 95%). This approach can certainly be transferred to stone and rock with similar porosities and hardness. Our approach also allows for consistent comparisons to be made across a wide variety of studies in the fields of rock weathering and stone deterioration research. Copyright © 2016 John Wiley & Sons, Ltd.