Heat transfer fluids play an important role in many industrial sectors. However, the low heat transfer characteristics of conventional fluids obstruct the performance enhancement and the high compactness of heat exchangers. In order to improve thermal characteristics of the conventional fluids, nanofluids are prepared by adding multi walled carbon nanotubes (CNTs) with base fluids. Though different experimental studies on nanofluids are available, theoretical models are also needed to predict its thermal behaviour. This work intends to address dimensional analysis using the Buckingham Pi theorem to develop an empirical model for predicting thermal characteristics of nanofluids. The latter will be achieved through the use of operational variables and physical properties for the identification of detrimental factors which eventually lead to the thermal enhancement of nanofluids. It can be observed from this analysis that volume fraction and temperature of the nanofluids are the most influencing parameters on the nanofluids thermal conductivity. In what concerns heat transfer coefficient, it is the velocity of the nanofluid that plays a critical role apart from the afore mentioned two parameters. Therefore it is believed that by controlling these parameters, the thermal effectiveness of the nanofluids can be established.