Introduction and background of the study

The relevance of environmental and social issues in the society—and especially in industrial activities—is calling national and international organizations, committees, and governments to develop a number of action plans and agreements aimed to increase sustainability at different levels (e.g., Kyoto Climate Change Protocol in 1997; COP21 Paris Agreement in 2015). Sustainability has been conceptualized by Elkington (1998) using the triple bottom line (TBL) model, as the intersection of three different pillars, namely, environmental, economic, and social. Focusing on an industrial context, we refer to industrial sustainability (Trianni et al. 2017), that it is related to all those actions that can be undertaken in a production plant level (and not just with reference to a production line) and that are referred to the levels of material, product, process, plant, and systems of production (Tonelli et al. 2013), and integrated into normal operations (Evans et al. 2009). Industrial sustainability has been often identified in literature with the areas of occupational health and safety (OHS) (Pagell and Gobeli 2009), and eco-efficiency (Gimenez et al. 2012), with a growing relevance of energy efficiency issue within eco-efficiency (Pehlken et al. 2015). Using the TBL model, we can identify these areas as the intersections of social and economic pillars (OHS), and environmental and economic pillars (eco-efficiency) (Pagell and Gobeli 2009; Gimenez et al. 2012).

To improve energy efficiency-related performance, it is necessary for firms to adopt energy efficiency measures (EEMs) (Rademaekers et al. 2011). Although there is good evidence that such measures are effective and have a positive impact on firms’ performance (Fleiter et al. 2012a), less than 50% of manufacturing firms have adopted EEMs (Anderson and Newell 2004; Cagno and Trianni 2012). Scholars have underlined the existence of barriers to energy efficiency improvement (Chiaroni et al. 2017). These barriers have been largely addressed in the literature, with both theoretical (Sorrell et al. 2000; Cagno et al. 2013) and empirical contributions. Regarding the former ones, scholars have studied barriers in different contexts such as firm sector (Henriques and Catarino 2016), country (Hassan et al. 2017), and firm size (Fresner et al. 2017); for a recent review of empirical studies, see, e.g., Brunke et al. (2014). Despite the deep investigation of barriers to EEMs, their adoption rate is still very low (Rasmussen 2014).

Cooremans (2011) suggested that EEMs are not adopted because they are not considered as strategic, i.e., able to create sustainable competitive advantages, and because no link is perceived between EEMs and firm’s core business. According to Cooremans (2012b), indeed, the mere increase in profitability, i.e., a financial analysis, is not enough to explain the low level of adoption, since several profitable measures are actually not adopted.

Nevertheless, literature has largely proven that adopting measures in the different areas of industrial sustainability, and in particular in the energy efficiency one, can improve competiveness and influence firm’s core business. Indeed, Lucato et al. (2017) affirmed that a pro-environmental attitude can increase competitiveness, while Das et al. (2008) stated that OHS-related measures lead to good quality management that in turn is linked to improvements in competitiveness (Gill 2009), as also confirmed in EASHW (2007). Regarding energy efficiency, Svensson and Paramonova (2017) purported that increasing energy efficiency is considered to be an important mean for increasing competitiveness, and the same is confirmed in McKinsey and Company (2012). According to other authors (Fleiter et al. 2012a), the strategic character of a specific EEM can be given in particular by non-energy benefits (NEBs). Indeed, according to IEA (2014), the multiple benefits can reveal the strategic value of energy efficiency, in terms of cost reduction, value increasing, and risk reduction (see also Cooremans 2011).

Several authors have suggested also considering the NEBs associated with the adoption of EEMs, i.e., those benefits related to the implementation of an EEM other than energy savings. NEBs can be looked as empirical evidence showing the impact of EEMs on other areas within the firm, and they can even amount to more than the energy savings (Pye and McKane 2000). A first categorization of NEBs was provided by Worrell et al. (2003) (the proposed categories are reduction of emission, material use, waste, time for maintenance; improvement of product quality, productivity, workers’ safety). Even if NEBs are well known, the authors underlined that firms lack the necessary knowledge to properly quantify them (Nehler and Ottosson 2014), and models for the quantification have been proposed (Ouyang and Ju 2017). An example of NEBs is provided, for instance by Trianni et al. (2014), according to whom an EEM related to the lighting may have also an impact on the working conditions, i.e., on safety issues. Nevertheless, on the one hand these relationships have been evaluated from an empirical viewpoint, and on the other hand, the different perspectives on the same EEM related to the different areas on which it may impact have not been studied, hitherto, in a holistic manner.

Hence, looking at EEMs and their barriers adopting an industrial sustainability point of view may help in better understanding all mechanisms lying behind the adoption of an EEM. Indeed, the presence of different perspectives (see also Cooremans 2012a; Thollander and Palm 2012) could provide added value to the comprehension of the problems related to adoption of EEMs, showing those so far hidden and helping in a more effective deployment of EEMs. Indeed, since the impact of the EEMs on the operations and on the other areas of industrial sustainability has been largely recognized, it would be interesting to broaden our perspective and understand if the issues related to the non-adoption of the measures can be related to industrial sustainability areas other than energy efficiency. For this specific purpose, the authors recently developed an integrated model for the evaluation of barriers to the adoption of measures to improve industrial sustainability performance (Trianni et al. 2017). Among those, for sure EEMs can be considered. Therefore, this model can be used by industrial decision-makers (IDMs) to evaluate barriers to the adoption of EEMs, pointing out possible sustainability issues hampering their adoption. Indeed, the model can identify general barriers to sustainability, as well as evaluate barriers to specific measures in the different areas of industrial sustainability (OHS, eco-efficiency, energy efficiency) and, therefore, could be very useful to understand problems related to the adoption of EEMs. In Table 1, we report the model with all barriers and their definition.

Table 1 The model on barriers to industrial sustainability
Full size table

Starting from this theoretical contribution, we aim to empirically investigate, on the one hand, the barriers to EEMs adoption from an industrial sustainability perspective and, on the other hand, the perspectives of the different IDMs knowledgeable about sustainability on the same EEM. Indeed, since an EEM affects several areas of the operations, multiple IDMs may influence its adoption. Hence, in our exploratory investigation, firstly, we are interested to understand whether different IDMs with different decision-making responsibilities in the different areas of industrial sustainability have different perceptions of barriers related to a specific EEM; secondly, beyond investigating whether possible positive reciprocal impacts among the different areas may support the implementation of an EEM, we would like to see whether and how the adoption of an EEM can be hindered by an IDM related to an area other than energy efficiency. Our analysis has been carried out through case studies conducted in 12 manufacturing firms located in Northern Italy.

The remainder of the paper is organized as follows: in the “Research methods” section, we present the theoretical framework used for the evaluation of barriers to EEM adoption and the research methods used for the empirical investigation (i.e., the case study methodology and the data collection and administration). In the “Results and discussion” section, we present and discuss our findings. Finally, conclusions are drawn and further research is suggested in the “Conclusions and further research” section.

Research methods

We have focused our exploratory empirical investigation on EEMs considered for implementation among manufacturing firms of Lombardy region (in Northern Italy), given its relevance for the Italian manufacturing sector and the still wide room for improvement in energy efficiency (ENEA 2016).

The empirical investigation is based on case study research methodology. This study fulfills the criteria for case study research identified by Yin (2009). We conducted the investigation through confirmatory case studies with semi-structured interviews, questionnaires, and secondary material. Twelve manufacturing firms differing in sector, size, and turnover were investigated (as shown in Table 2), following previous research pointing out that investigating a heterogeneous sample of firms provides evidence for the generalizability of an emerging theory (Eisenhardt 1989). Considering the need to judge the theoretical generalizability of the research (Hillebrand et al. 2001; Stuart et al. 2002) rather than its statistical generalizability, our number of selected case studies is deemed to be enough to provide valid support for the initial set of propositions (Eisenhardt 1989; Pagell and Wu 2009), allowing also depth of observation (Zorzini et al. 2008). To ensure that we collected appropriate data, with the aim of predicting similar results from the case studies (Shakir 2002), we identified interviewees able to provide specific information regarding EEMs and their impact on the operations and firm sustainability (Voss et al. 2002). Therefore, we selected in each firm people knowledgeable and responsible for energy issues (i.e., energy efficiency), environmental issues (i.e., eco-efficiency), and safety issues (i.e., OHS). We interviewed 24 people in charge of energy efficiency, eco-efficiency, and OHS within the sampled firms, ensuring to have at least two managers in each firm, so to compare different perspectives, e.g., interviewees from energy and environmental area, and from OHS area. We interviewed each manager separately to better capture the personal judgments and frank opinions, thus limiting as much as possible any bias due to, e.g., different power within the firm (for further detail see (Eisenhardt and Zbaracki 1992)). We developed a case study protocol for helping us standardize the sequence in which the questions were asked and minimize the impact of contextual effects (Patton, 1990). Each face-to-face interview lasted approximately 2 h.

Table 2 Data of the investigated firms. For each firm, the sector, a short description of the activity, the number of employees, turnover, certifications owned and managers interviewed are reported
Full size table

The data collection has been organized in three parts. The first corresponded to the identification of the research sample using a database (AIDA 2017) containing relevant industrial information. Firms were selected basing on sector, number of employees, turnover, and geographical location. Firms were contacted by e-mail or phone call and, for all those that accepted to participate to the research, secondary firm data (firm websites, reports, newspapers) were collected, regarding firm structure, production processes, their (where available) projects, initiatives and similar, towards increased industrial sustainability.

The second part corresponded to the investigation within the sampled firms. Each investigation was performed adopting semi-structured interviews, audio-recorded and transcribed for analysis, with a questionnaire used as a guide, so to standardize the sequence in which the questions were asked and minimize the impact of contextual effects (Patton 1990). We based the interviews around a series of open-ended questions, which were supplemented by questions emerging from the dialog between the interviewer and interviewees, and probes (Remler and Van Ryzin 2014). We also collected free comments, in line with the procedure described by Dicicco-Bloom and Crabtree (2006). To start, each interviewee was asked to introduce the firm to the interviewer (i.e., sector, production process, number of employees, turnover, and attitude towards sustainability). This allowed to have a first corroboration of the data found in the web and to ask interviewee to explain possible misalignments, in particular regarding their attitude towards sustainability. The first manager interviewed in each firm was asked to arrange a tour of the plant for the interviewer. This allowed the interviewer to directly observe and evaluate how the plant worked and to identify possible problems related to industrial sustainability areas. After the tour, the interview took place. We presented the model of barriers to each interviewee, describing every single barrier. Interviewee was provided with a list of industrial sustainability measures (we adopted the one proposed by Trianni et al. 2017) and asked to identify, among the measures, those that were considered for adoption within their firm. For these measures, the interviewee was asked to evaluate, using the model proposed, the main barriers faced for their adoption and to discuss possible additional measures missing from the list. For each measure considered for adoption, the interviewee was asked to recount the whole decision-making steps followed, contextualizing the situation in which the adoption took place and to explain in detail the impact of that barrier in the specific situation. Main insights and issues that emerged from the evaluation of barriers were further investigated. The interviewee was then asked to rate the relevance of barriers using a four-point Likert scale, where 1 is “not relevant,” 2 is “low-medium relevance,” 3 is “medium-high relevance,” and 4 is “high-very high relevance.” Using a Likert scale to collect data on the relevance of barriers enabled us to synthesize the data from all interviewees and provide a quantitative measure, thus supplementing the comments and evaluations. An even four-point Likert-like was chosen, so as to push the respondents into taking a position, as done by previous research (Massoud et al. 2010; Fleiter et al. 2012b).

The third part of the data collection corresponded to the transcription and coding of the interviews and to the identification of possible misalignments that emerged, identified through the corroboration of the data obtained from the different sources (i.e., semi-structured interview, tours of plants, Likert-like scale, secondary data). In case of misalignments, we called back the interviewees, asking for a second face-to-face meeting or a phone-arranged one, in order to clarify these misalignments.

According to Yin (2009), four requirements must be met to guarantee the methodological rigor of case study research.

First, construct validity is the establishment of operational measures, obtained with triangulation of multiple source of evidence and with the development of a chain of evidence. Regarding triangulation of multiple sources of evidence (Voss et al. 2002; Beverland and Lindgreen 2010), in our investigation, we corroborated the data obtained using semi-structured interviews, direct observations, and secondary material, i.e., company’s report and websites (Baškarada 2014). Concerning the chain of evidence, this is considered necessary to understand how the researchers arrived at their research outcomes from the data that was collected (Benbasat et al. 1987); basing on Rowley (2002) for every firm investigated, we create an electronic folder containing secondary data with related notes, interview transcript, notes taken during the interview and during the tour of the plant, and coding of the interview. Regarding the coding, we used structural coding since it is considered appropriate for exploratory semi-structured investigation in which multiple participants are involved (Saldaña 2009), and the main themes used were strictly related to the research questions of the study, i.e., barriers to the adoption of EEMs and different perspectives on them according to the different IDMs.

Second, internal validity is the extent to which casual relationships can be established: according to Yin (2009), Beverland and Lindgreen (2010), and Baškarada (2014), it only applies to explanatory and not to descriptive or exploratory case studies.

Third, external validity is the extent to which results can be generalized; this was assessed by defining the domain to which study findings can be generalized, i.e., the specification of population, replication logic, and the use of multiple case studies (Beverland and Lindgreen 2010).

Fourth, reliability is concerned with demonstrating that same results can be obtained by repeating the data collection procedure; it was addressed with the use of a case study protocol (Beverland and Lindgreen 2010) that standardizes the investigation and with the creation of a case study database.

In order to eliminate possible researcher bias, on the one hand, multiple case studies were conducted (Barratt et al. 2011), and on the other hand, more than one interviewers were involved in each interview and each interview was tape recording, as suggested by Voss et al. (2002).

Results and discussion

The investigated EEMs for each firm have been reported in Table 4. Each EEM has been categorized according to its main impact on the different areas of industrial sustainability. For each measure, we reported, where present, barriers with a value equal to or greater than 3 of the Likert-like scale. We also provided further comments regarding the implementation of the EEM. In the following, the discussion is structured according to the main research issues addressed in the study.

Existence of multiple perspectives on barriers to energy efficiency measures

During our exploratory investigation, we observed that the different IDMs of the industrial sustainability areas may have different perspectives on the same EEM, as well as perceive different barriers on their adoption, as can be inferred from Table 4. In particular, the existence of multiple perspectives on barriers to EEMs has been observed in all the firms investigated. In eight firms out of 12, this has been observed even in most of the EEMs discussed. The second column of Table 3 summarizes the findings for this point.

Table 3 Result. The table reports the summary of findings in each investigated firm
Full size table

In firm A, OHS manager was totally underestimating barriers to the adoption of EEMs, with respect to energy and environmental manager. For each EEM proposed, the first identified almost no barriers for its implementation, stating that, in general, EEMs were implemented without any problem. In contrast, the latter identified several barriers, particularly related to a general attitude of the organization (because of other priorities and lack of awareness), to a lack of proper information, to a lack of time, and to economic barriers. Moreover, the investigation showed, beside a different view on the barriers, a different knowledge of IDMs regarding the implementation of EEMs. The OHS manager stated that, e.g., preventive maintenance was not carried out, as he asserted they “do not have specific weekly or monthly commitment for preventive maintenance,” and maintenance activities were implemented only after a machine failure; on the contrary, the energy and environment manager pointed out that a maintenance team should have periodically controlled the machines and that, although these activities were scheduled, very often they were not implemented due to lack of time and the costs related to the production disruption. Moreover, workers should have implemented preventive maintenance during their working hours, but, as energy and environment manager stated, “in this way they have to interrupt their normal activities, postponing them, or have to stay at work after the normal working hours,” adding that preventive maintenance “is perceived by workers as a waste of time.”

In several other cases, we detected that OHS managers were often unaware of barriers related to the adoptionof EEMs. For example, firm D implemented the EEM “energy efficiency training” once per year after the achievement of ISO 14001 certification. Managers tried to further involve workers in energy efficiency issues by asking them to provide suggestions and advice, as energy efficiency manager said “workers can suggest possible actions to be undertaken so to improve energy efficiency: there is a PO box in the industrial building and everyone can write a mail with suggestions.” OHS manager did not pinpoint any relevant barrier, underlining that training was strongly supported by top management, whereas energy and environment manager pointed that, in daily activities, possible positive effects of training on production were nullified by incorrect behavior of workers. Another example is the substitution of existing lamps with more efficient ones in firm E. Both managers recognized the investment costs as a main barrier, and they highlighted that, for this reason, the EEM was only partially implemented. The energy and environment manager however further explained that this barrier was related to the management’s inability to see future benefits from the implementation of that EEM (e.g., savings) and thus a lack of a long-term vision. He also related this situation to a resistance to change.

Finally, in some cases, different IDMs of industrial sustainability areas not only agreed on the relevance of barriers to the adoption of a specific measures but also recognized the existence of an additional perspective (i.e., the top manager’s one) hindering the adoption of the EEM. Installation of extractor fans, indeed, was strongly supported by both managers in firm L. Born as a measure for improving workers’ comfort, both managers recognized it as being able to bring energy savings to installed equipment. Despite the existence of a feasibility study showing the opportunity to have energy savings and improved working conditions, as well as the positive evaluations from both managers, the management decided to perform a test by installing only two extractors out of the six proposed and to evaluate the positive effects deriving from this installation. By limiting the scope of the EEM, the management was not able to effectively experience the full set of expected benefits after the installation, so he decided to stop a further investment in the EEM. In this case, the management, indeed, showed to be unable to properly assess benefits derived from the EEM adoption. The OHS manager in particular pointed out that: “the benefit deriving from the control of the temperature related to the installation of the fans would have been twofold. Indeed, when there are more than 25°C in the production department, on the one hand, workers start to feel tired more easily and their level of attention is low; on the other hand, machines go into crisis, the process becomes longer and the energy consumption increases.”

Our exploratory investigation preliminarily shows that, for different IDMs related to the different areas of industrial sustainability, different perspectives on the relevance of the barriers to the implementation of an EEM may exist. This finding is in line with the research by Langley et al. (1995) that emphasizes the individual rather than the organizational level of analysis of the decision-making process, underlying how the process is mainly driven by personal insights and emotions. As a consequence, in order to have a more thorough comprehension of the barriers affecting EEMs, it seems quite beneficial to broaden the perspective, thus enlarging from an energy efficiency to an industrial sustainability one. Indeed, during the analysis of barriers to EEMs, our study revealed that considerable other information can be inferred from other IDMs’ perspectives beyond the energy-related one. This is even more interesting for giving a proper boost to the adoption of EEMs. In fact, if IDMs referring to other areas of sustainability are unaware of existing barriers to EEMs, they could not provide a valuable support for its effective implementation. For this reason, considerations regarding the involvement of energy managers at top level of a company’s organizational chart (see, e.g., Sorrell et al. 2010; Thollander and Palm 2015) are really crucial for the promotion of energy efficiency and sustainability in industrial activities, as it has been largely recognized that the characteristics of the management (including beliefs, theories, and propositions based on managers’ personal experience) are critical for explaining the performance of a firm (Prahalad and Bettis 1986; Bettis and Prahalad 1995). Indeed, it is important to give energy manager power influence, i.e., provide them with formal authority, control of scarce resources (i.e., skills and money), and information and knowledge: indeed, basing on the assumption that firms are coalitions of people with competing goals coming from their positions within the firm and personal ambitions and interests (Eisenhardt and Zbaracki 1992), the project champions very often do not succeed because they struggle in overcoming barriers created by divisional structure (Sorrell et al. 2000; Masi et al. 2014). In particular, the complexity of the decision-making process for sustainability-related decision has been largely underlined (Gibson 2006; Arvai et al. 2012), and it has been related to the presence of trade-offs among the performances concerning different pillars of sustainability, the time span considered (short, medium, long), and the different stakeholder requirements (Nicolăescu et al. 2015; Gong et al. 2016; Frini and Benamor 2017).

Energy efficiency measure adoption can be negatively affected by other areas of sustainability

In our exploratory investigation, frequently, the implementation of an EEM was positively or negatively affected by reasons related to other industrial sustainability areas within the firm, as can be inferred from Table 4. Regarding EEM adoption affected by other areas of sustainability, in 6 cases out of 12, it was possible to observe that EEM adoption was positively affected by other areas of industrial sustainability, but more relevant was to observe that in 5 firms out of 12, EEM adoption was negatively affected by other areas of sustainability. The third column of Table 3 summarizes the findings for this point.

Table 4 The EEMs discussed during the interviews. For each firm investigated, we listed the EEMs considered during the interviews. Each measure is categorized according to the impact on the different areas of industrial sustainability (EnEff: energy efficiency; EcoEff: eco-efficiency; OHS: occupational health and safety), according to Trianni et al. (2017). Managers interviewed and main barriers identified by each of them (Likert-like scale value ≥ 3) are reported and further comments regarding the specific EEM are provided
Full size table

We detected that positive reciprocal impacts may exist between energy efficiency area and the other industrial sustainability areas of the firms. In particular, EEMs may have positive effects on other areas, and measures originally related to other areas, such as safety, may have positive effects on energy efficiency. For instance, the substitution of existing lamps with more efficient ones proved to bring safety-related benefits in more than one firm. Such benefits can be as in, e.g., firm C, improvement of workers’ comfort and the reduction of power, and, as a consequence, the reduction of absorption, dissipated power, voltage drops, and danger. Furthermore, the installation of combined heat and power system in firm I for substituting the previous heating system allowed to reduce the energy consumption and costs associated with heating and to eliminate the electrical resistances needed by the previous system, thus avoiding the concrete possibility of risk of a fire: indeed, as the environmental and safety manager said, they “used to have a heating system with resistances inside, that, for an error, went in short-circuit and caused an initial fire.” Finally, the installation of glass roofing in some parts of the production plant in firm L to reduce the need for artificial illumination and use daylight as much as possible also brought benefits related to working conditions, in particular to comfort.

Interestingly, we also ascertained new with respect to previous literature that safety-related measure brought energy efficiency-related benefits. This occurred in firm K, in which original brick walls of the production departments were painted white to make the space brighter and improve workers’ comfort. Even if this measure was primarily aimed at increased safety, the firm also experienced energy benefits. Indeed, with a brighter space, the need for lighting was reduced, with positive impact in terms of energy and economic savings, as health, safety, and environmental manager said: “we implement this measure for reasons not related to lighting […] but it turned out to benefit lighting and so energy consumption”.

We detected that EEM adoption may be hindered by reasons related to other areas of industrial sustainability. As from our investigation, this negative impact can be observed according to factors as follows. Firstly, workers’ comfort prevailed over energy firm performance. For instance, firm A moved a machine to a place in which fewer workers operate and with a higher ceiling, in order to more easily disperse the noise. Despite the change and the low use of the machine (about only 1 day every 2 weeks), some processing parameters were lowered to reduce the perceived still loud noise, with negative impact on production performances of the machine, and increased energy consumption. In this case, as energy and environment manager revealed, “workers were properly equipped with ear protections, but they did not use them. Nevertheless, they complained about the noise and, to guarantee a comfortable place for workers to work in, it was decided to lower the parameters.”

Secondly, similarly to what was shown by Trianni et al. (2013), other priorities may lower the urgency of EEMs, such as interventions that guarantee compliance with safety regulations and allow a firm to continue its production activity. For example, in firm B, the substitution of existing lamps with more efficient ones was recognized as particularly critical by both managers. Firm B had asbestos in the roof that should have been removed years before. Nevertheless, top management had so far postponed the decision, because of the extant opportunity to move to another plant. Eight years later, on the one hand, the firm had not moved yet; on the other hand, so far, no interventions had been implemented on the roof. But, at the time of the interview, the firm experienced several structural problems in the roof and had to remove the asbestos due to regulatory issues. In a nutshell, despite the positive evaluation of both OHS and energy, maintenance and environment managers (the first even stating “it has been ten years since I proposed to change the lighting”), now, the priority of regulatory (safety) issues emerged, stopping any further investment in energy efficiency. In particular, the energy, maintenance, and environment manager clearly stated: “at this moment, all those interventions that are not included in the building revamping are not considered” and “we privilege those interventions that keep us alive, rather than those that give us an economic benefit”.

The aforementioned considerations seem to point out that the set of performances of an EEM to be taken into account when adopting it goes beyond the energy efficiency ones. In fact, our empirical evidence shows that firms cannot avoid safety and comfort issues when implementing EEMs. Positive reciprocal impacts among the different areas, indeed, may support the implementation of an EEM. EEMs can be positively affected by reasons related to other areas of industrial sustainability, in particular, findings underlined strong relationships with OHS area. In this way, NEBs may foster the implementation of EEMs, confirming previous literature that pointed out possible benefits stemming from the adoption of EEMs (Morrow et al. 2014; Nehler and Rasmussen 2016). It has also emerged that energy efficiency reasons may positively affect the adoption of measures related to other areas of industrial sustainability, so that energy benefits may foster the implementation of non-energy measures. In the same way, EEMs can be hindered by an IDM related to an area other than energy efficiency. From the investigation, a strong relationship with the OHS area emerged. Indeed, EEMs can be stopped for reasons related to safety that concern, e.g., workers’ safety and comfort or the need to be compliant with safety regulations. Firms cannot avoid such aspects when implementing EEMs. Nevertheless, too little attention has been so paid hitherto to analyze the negative consequences that may arise from the implementation of an EEM (Trianni et al. 2017), thus extending the perspective on industrial sustainability beyond energy efficiency performance.

Conclusions and further research

There is a growing concern (Omer 2008; Dincer and Rosen 2012) regarding the adoption of EEMs as relevant contributors to industrial sustainability. Through our exploratory investigation, we have empirically shown that looking at EEMs and their barriers adopting an industrial sustainability point of view may help in better understanding all those mechanisms lying behind the adoption of an EEM, hinting that the presence of different perspectives is able to provide added value to the comprehension of the problems related to adoption of EEMs. Indeed, our investigation revealed that different IDMs seem to have different perspectives on the relevance of the barriers in the adoption of a specific EEM. This, of course, impacts on the adoption itself and a more proper evaluation of all the issues related to the adoption seem possible broadening the perspective, from an energy efficiency to an industrial sustainability one. Furthermore, our sample pointed out that, if in some cases the EEM adoption may have positive reciprocal impacts with other areas of industrial sustainability, in other cases, EEMs can be negatively affected by reasons related to areas others than the energy efficiency one. Stemming from the obtained findings, it is possible to conclude that, when adopting an EEM, it is necessary to consider not only the energy area but also all those areas that may be involved in the implementation of an EEM, i.e., to broaden the perspective towards an industrial sustainability one, so as to have a more complete and proper view on all those factors that may hinder or foster the adoption of an EEM. It becomes clear, indeed, that if we really want to increase the rate of adoption of EEMs, it is necessary to consider all their impacts and thus all the different perspectives related to them. On the one hand, the perspectives that IDMs related to of all industrial sustainability areas may have about the EEM should not be overlooked; on the other hand, for the effective implementation of an EEM, it is important to take into consideration the impact of the EEMs on other areas of industrial sustainability.

Our findings may offer relevant suggestions to IDMs as well as policy makers in order to, on the one hand, point out the best drivers to tackle existing barriers and, on the other hand, identify the most suitable stakeholders within the firm (or outside) to promote such drivers. The results obtained would also be useful for technology/service suppliers, i.e., properly identifying in the firm their right counterparts for the promotion of their products/services within the firm.

Despite that the study provides a good empirical validation of the initial set of propositions, it presents some limitations, that howbeit has offered the opportunity to sketch some future research. First, we were not able to interview people in exactly the same leadership position among the different firms. Moreover, the results obtained provide only a theoretical generalizability of the results. Further research may, for sure, enlarge the sample. This would allow having a statistical generalizability too, investigating possible common patterns, i.e., according to firms’ clusters related to their characteristics and contextual factors, such as, e.g., geographical area, sector, dimension, energy intensity, types of processes, organizational structure.

In addition, further research could understand the role of energy efficiency in preventing or supporting the implementation of measures related to the other areas of industrial sustainability. Both for EEMs and for measures related to other areas of industrial sustainability, it would be interesting to analyze together main barriers and main drivers related to their adoption and to evaluate their relevance according to multiple perspectives related to different IDMs knowledgeable about industrial sustainability. Furthermore, to offer a valuable support to IDMs as well as policy makers in the promotion of sustainability measures, it would be quite important to link the adoption of EEMs to the broad set of sustainability performance. For this reason, further research could explore the relationships that, with respect to a specific measure, exist among barriers, drivers, level of adoption of the measure, and sustainability performance reached. Such type of analysis should not be necessarily limited with the boundaries of a single firm. Indeed, future research could analyze such relationships according to the different perspectives of different firms belonging to, e.g., the same supply chain and industrial district.