Enhanced optical absorption of molecules in the vicinity of metallic nanostructures is key to a number of surface-enhanced spectroscopies and of great general interest to the fields of plasmonics and nano-optics. Yet, experimental access to this absorbance has long proven elusive. We here present direct measurements of the intrinsic absorbance of dye-molecules adsorbed onto silver nanospheres, and crucially, at sub-monolayer concentrations where dye-dye interactions become negligible. With a large detuning from the plasmon resonance, distinct shifts and broadening of the molecular resonances reveal the intrinsic properties of the dye in contact with the metal colloid, in contrast to the often studied strong-coupling regime where the optical properties of the dye-molecules cannot be isolated. The observation of these shifts together with the ability to routinely measure them has broad implications in the interpretation of experiments involving resonant molecules on metallic surfaces, such as surface-enhanced spectroscopies and many aspects of molecular plasmonics.Over the last two decades, the optical properties of metallic nanoparticles (NPs) of various sizes and shapes [1] [18][19][20]. Many of these existing and emerging applications are underpinned by the fact that the optical (electronic) absorption of molecules on the surface of metallic NPs is enhanced. But spectral changes induced by molecular adsorption are often ignored because of the experimental challenge of measuring surface absorbance spectra on nanoparticles, despite early attempts more than 30 years ago [21].This question is not directly addressed in the great number of recent studies devoted to the topic of strongcoupling between plasmons and molecules [22][23][24][25][26][27][28][29][30][31][32][33][34][35][36]; in this regime, the plasmon-molecule interaction is evidenced by a typical anti-crossing of the two resonances as a function of detuning [25,33], but in such a strongly interacting system the molecular response cannot be isolated. Moreover, in such studies the dye concentration is often large (typically monolayer coverage and above) to maximize dye/plasmon interactions. Dye-dye interactions cannot therefore be neglected and are expected to induce resonance shifts of the dye layer independently of any plas- * Eric.LeRu@vuw.ac.nz monic effects; in fact many studies specifically work with J-aggregates rather than isolated dyes [22, 25-27, 31, 33]. As a result, the intrinsic effect (chemical and/or electromagnetic) of the NP on an isolated adsorbed molecule cannot be elucidated. To this aim, we will show that it is necessary to measure the absorbance of the dye when the dye and plasmon resonances are detuned (to avoid dye-plasmon interaction effects) and at low surface coverage (to avoid dye-dye interaction effects). This low surface coverage poses a significant experimental challenge because the dye absorbance is then very small and obscured by the large optical response (absorption and scattering) of the NPs.We here propose to measure NP/d...