A metal dimeric nanostructure is an alternative design for a Localized Surface Plasmon Resonance sensor. At the dimer nanojunction, a high E-field enhancement (hot spot) is observed. In this work, nanostructure size parameters such as aspect ratio, junction, and length of gold dimeric nanorods (AuDNR) linked end-to-end by a thin dielectric junction were evaluated numerically, optimizing the sensing parameters of the plasmonic nanoplatform. The computational approach focused on the assessment of the LSPR extinction spectral peak shift to their local surrounding medium in order to estimate the behavior of sensing parameters such as the figure of merit (FoM), the bulk sensitivity ( η <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">B</sub> ) and the η <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">B</sub> × FoM. The nanostructure aspect ratio, junction, and rod length were evaluated, optimizing the plasmonic nanoplatforms sensing parameters. The optimized AuDNR achieved a high sensing performance with η <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">B</sub> = 544 nm/RIU and FoM = 12.4 RIU <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-1</sup> . Moreover, the obtained η <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">B</sub> × FoM = 6745.6 nm/RIU <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> is the highest reported value in the literature. For molecular sensing, Campbell's model was explored to evaluate the wavelength shift as a result of molecular adsorption on the nanoparticle surface. We demonstrate that the dimeric nanostructure exhibit twice LSPR peak shift due to a thin molecular adsorption layer rather than a monomer of equivalent size. Our results show that optimized dimeric nanostructure is highly attractive for molecular identification, opening the possibility of establishing new nanoplatforms for medical diagnostic.