Adhesion is an intrinsic property of rocks and liquids. Investigating the factors contributing to its formation and the mechanisms governing its action is crucial for elucidating the adhesion work between solids and liquids. The adhesion work, serving as a parameter that characterizes the energy changes during the solid−liquid contact process, is a vital tool for probing this phenomenon. However, conventional measurements of the adhesion work are significantly influenced by surface roughness and fail to differentiate local variations in the adhesion performance. This limitation obscures our understanding of the primary adsorption sites and mechanisms between solids and liquids, posing significant challenges to the study of rock surface properties. In this study, in conjunction with scanning electron microscopy and contact angle analyses, we elucidated for the first time the locations where voids form during the solid−liquid contact process, the lithological composition of rough areas, and their impact on the adhesion work between water/oil and the surfaces. Additionally, employing atomic force microscopy (AFM), we examined the variations in water/oil−solid adhesion work across different characteristic regions, thereby characterizing the overall hydrophilic/ hydrophobic properties of the rock core. Specific conclusions are as follows: (1) A negative correlation exists between roughness and the contact angle adhesion work, with heterogeneity impeding liquid-rock contact; (2) By comparing the strength of water−solid/ oil−solid adhesion work within localized areas, we delineated the adhesion work characteristics of samples and their primary generation sites, with oil−solid adhesion work in target blocks predominantly originating from quartz, clay minerals, and organic matter; (3) The influence of pore throat development on the overall adhesion work of samples was clarified, demonstrating that an increase in the proportion of internal rock pores enhances the surface oil−solid adhesion work; (4) A dimensionless wetting index I was established to mitigate the impact of roughness on the expression of adhesion work, exhibiting a strong correlation with traditional evaluation methods.