In a scenario characterised by mitigation concerns and calls for greater resilience in the energy sector, energy audits (EAs) emerge as an essential mean for enhancing end-use energy consumption awareness and efficiency. Such a tool allows us to assess the different energy carriers consumed in a productive sector, offering insight into existing energy efficiency improvement opportunities. This opens avenues for research to devise an econometrics-based methodology that encapsulate production sites and their environmental essentials. This paper contributes to the literature by exploiting the EAs received by the Italian National agency for New technologies, Energy, and Sustainable Economic Development (ENEA) in 2019 from the Italian plastics manufacturing sector, matched with Italian firm-based data extracted from the Analisi Informatizzata delle Aziende Italiane (Italian company information and business intelligence) (AIDA) database. In particular, we investigate how the implementation of energy efficiency measures (EEMs) is influenced by a set of contextual factors, as well as features relating to the companies and EEMs themselves. The empirical investigation focuses on the EAs submitted to ENEA in 2019, which was strategically chosen due to its unique data availability and adequacy for extensive analysis. The selection of 2019 is justified as it constitutes the second mandatory reporting period for energy audits, in contrast to the 2022 data, which are currently undergoing detailed refinement. In line with the literature, the adopted empirical approach involves the use of both the OLS and logistic regression models. Empirical results confirm the relevance of economic and financial factors in guiding the decisions surrounding the sector’s energy performance, alongside the analogous influence of the technical characteristics of the measures themselves and of the firms’ strategies. In particular, the OLS model with no fixed effects shows that a one-percent variation in investments is associated with an increase in savings performance equal to 0.63%. As for the OLS model, including fixed effects, the elasticity among the two variables concerned reaches 0.87%, while in the logistic regression, if the investment carried out by the production sites increases, the expected percentage change in the probability that the energy-saving performance is above its average is about 187.77%. Contextual factors that prove to be equally influential include the incentive mechanism considered and the traits of the geographical area in which the companies are located. Relevant policy implications derived from this analysis include the importance of reducing informational barriers about EEMs and increasing technical assistance, which can be crucial for identifying and implementing effective energy solutions.