Achieving social impact through business research aimed at contributing to climate change mitigation Open Access

The social impact of research has gained considerable attention in recent years, particularly in the field of business research, which increasingly investigates the importance of social impact and strategies for generating research outputs with high social impact (Rawhouser et al., 2019). However, creating socially impactful research remains a challenging task. This is due to the complex nature of social impact, which includes the overall assessment of research outcomes across four key domains: social, cultural, environmental and economic benefits (Bornmann, 2013). Social impact is also inherently longitudinal, with an emphasis on how research results generate mid- to long-term benefits at both the individual and community levels (Ruegg and Feller, 2003). As a result, many attempts by business researchers have failed to create the desired social impact (Parkinson and Naidu, 2024). The current landscape, therefore, seems to be characterised by business researchers knowing that impact is a key dimension of their work, but they do not know how to achieve impact.

This study aims to address this gap by proposing a practical framework to guide business researchers in translating their theoretical knowledge into practical outputs that deliver direct social impact. The framework we propose can be applied to any of the four social impact dimensions. In this article, however, we focus on environmental benefits because of the existential climate change challenge humanity is facing.

Environmental sustainability has gained considerable attention in recent decades because of the climate emergency and because of the spill-over effects to other areas of social impact (Taghvaee et al., 2023). The trend of recycling, for instance, primarily generates environmental benefits, but also spills over to social benefits (e.g. improved recycling standards and techniques; Ding and Zhu, 2023), culture benefits (e.g. cultivating a sustainable lifestyle; Ha et al., 2023) and economic benefits (e.g. promoting the circular economy; Neumann et al., 2022). Environmental sustainability, therefore, represents a key area of social impact. The importance of environmental sustainability is also reflected at the national and international policy level, including in the United Nations Sustainable Development Goals and the net-zero goals for 2030 (United Nations, 2015).

The importance and urgency of addressing climate change presents opportunities and challenges for both practitioners and business researchers. Although climate mitigation strategies are developing to become part of the default operational agenda for many large companies (e.g. Amazon and Hilton), small- and medium-sized organisations are lagging behind because they lack knowledge and face financial constraints (Purwandani and Michaud, 2021). It is, therefore, important for business researchers to focus on small- and medium-sized organisations and provide them with solutions to improve their environmental performance. With small- and medium-sized organisations representing about 90% of businesses and offering more than 50% of employment worldwide (World Bank, 2023), this could arguably be the most effective way to create maximum social impact. Yet, most business research focuses on theory development, with little practical guidance for businesses.

This article proposes a framework for generating social impact in business research, with environmental sustainability as the core driver. From a systems perspective, improvements in environmental performance often produce positive spill-over effects to other domains – such as enhanced social equity, collective wellbeing and societal resilience (Taghvaee et al., 2023). This perspective aligns with a multidimensional framework, which conceptualises impact as comprising environmental, social, cultural and economic aspects (Bornmann, 2013). The UN 2030 Agenda for Sustainable Development also emphasises that environmental goals, including SDG 13 (Climate Action) is the foundation for achieving boarder societal objectives. We therefore argue that addressing climate change mitigation through applied business research can generate not only positive environmental benefits, but also socially transformative outcomes.

We first provide an overview of business research on climate change mitigation, then explain Mode 1–Mode 2 Theory (Gibbons et al., 1994), which serves as the theoretical basis of our framework. Next, we introduce and discuss the proposed new framework, which contains five building blocks for business research to achieve social impact. At each step of this framework, we discuss in detail – and illustrate with actual business research – examples with environmental benefits.

Business research for climate change mitigation covers a wide range of topics. Environmental accounting research focuses on the financial implications of carbon footprint reduction strategies (He et al., 2022); environmental marketing research focuses on the promotion of environmentally sustainable products or services (Gelderman et al., 2021); and green supply chain research aims to reduce waste and emissions throughout the supply chain (Yadav et al., 2021). Despite different research foci, all these areas of business research aim to contribute to climate change mitigation. However, the extent to which research findings can generate social impact depends less on the specific research topic, and more on the study’s primary focus – whether it aims to advance theoretical understanding or to create practical outcomes (Bornmann, 2013).

With the trend towards evidence-based decision making, there is also an increased expectation for business research to draw conclusions that are valid and reliable and, as such, can be used as a solid basis for decision-making for maximum social impact (Martin, 2011). This shift in the research landscape has been described as a move from “Mode 1” to “Mode 2” research (Gibbons et al., 1994). This Mode 1–Mode 2 Theory guides the development of our framework. The theory posits that Mode 1 research is governed by the interests of the academic community, typically a focus on pushing theoretical boundaries of knowledge (Walter et al., 2007). In the context of business research for climate change mitigation, Mode 1 research includes developing theories on behavioural change (e.g. Theory of Planned Behaviour; Ajzen, 2011; Dolnicar et al., 2023a, 2023b) and defining emerging concepts (e.g. sustainability indices; Mori and Christodoulou, 2012).

Mode 1 research creates the foundations for business research. It has contributed significantly to knowledge development (Bornmann, 2013) but has attracted criticism since the 1990s (e.g. Salter and Martin, 2001) because Mode 1 research fails to offer practical guidance. This criticism is particularly relevant to environmental sustainability research, which aspires to achieve carbon emissions reductions – a highly practical outcome. As a result, the focus of business research has shifted towards Mode 2 research. Mode 2 research focuses on applying basic knowledge to solve practical problems; it is often characterised by collaboration between the scientific community and industry to solve a particular problem or overcome a particular challenge (Ernø-Kjølhede and Hansson, 2011). Compared to Mode 1 research, Mode 2 research has the potential to achieve greater immediate social impact via industry collaboration (Bornmann, 2013); results from Mode 2 research are evaluated based on utilitarian values and criteria (Petit, 2004). The impact of Mode 2 research, therefore, needs to be assessed using real-world data to measure the practical value. In the context of environmental sustainability, this can be measured through explicit behavioural measures, which can then be converted into environmental indicators (e.g. carbon footprint) and economic benefits (e.g. savings). Measuring actual behaviour – rather than relying on stated behavioural intentions – is a key characteristic of Mode 2 research. Importantly, moving from Mode 1 towards Mode 2 research does not imply that theoretical knowledge development will be compromised. Measuring actual behaviour helps researchers test and refine existing theoretical knowledge, thereby advancing theory even when the behavioural change intervention fails. For example, tourism researchers have found that belief-based interventions are often less effective in promoting environmentally sustainable tourist behaviour, suggesting that the hedonic nature of holiday may require alternative theoretical foundations for behavioural change (Demeter et al., 2023). Optimally, Mode 2 research still pushes the boundaries of knowledge, while additionally producing outcomes that can be adopted to achieve social impact.

Mode 1 research dominates work on climate change mitigation (Arteaga et al., 2023). Guidance for business researchers wanting to translate their academic insights into practical solutions that achieve impact is not readily available. Building on Gibbons et al.'s (1994) Mode 1–Mode 2 Theory, we develop a framework that aims to offer such guidance (Figure 1). In this section, we discuss the role of each building block of this framework and how researchers can achieve social impact of their research based on this framework. Maximising social impact does not mean that the translation is binary – from no impact to maximum impact. Rather, every step towards more impact is a positive outcome. The framework, thus, aims to offer guidance on how impact can be increased, without dictating the magnitude of the change. The building blocks of the framework are intended to serve as a practical guide for researchers to enhance social impact. Building blocks do not need to occur sequential, rather each one of them could serve as entry point into a project that generates social impact. The ability of engaging in Mode 2 research depends on external factors such as funding availability and level of industry engagement.

Any Mode 2 research that aspires to solve a practical problem and, in so doing, achieve real-world impact, benefits from leveraging fundamental theory and knowledge. For example, when aiming to change a specific behaviour, designing a behaviour change intervention based on our collective theoretical understanding of what drives that behaviour increases the likelihood of the intervention being effective. Mode 2 research is not meant to be a trial-and-error exercise. Rather, Mode 2 research builds on theoretical and foundational knowledge (Mode 1 research) and translates it into solutions for real-world challenges.

In most instances, measuring actual impact requires a field study to be conducted. Field studies are becoming more common, but are still outnumbered by survey studies, survey experiments and laboratory experiments, which cannot achieve the required external validity (e.g. Dolnicar et al., 2024b). As an extension of Figure 1, Figure 2 provides a more detailed explanation of how impact is tested in field studies. A field study aims to determine the impact of a specific intervention in the following way: first, a pre-intervention baseline of the behaviour of interest is recorded. This involves measuring, typically over an extended period, the target measure (dependent variable). Typical measures in business research focusing on climate change mitigation include electricity used, water used, waste generated or, where available, carbon emissions generated. Baseline measurements must be taken over an extended period because of variability in the data caused by external factors (intervening variables). For example, if the aim is to reduce food waste at a breakfast buffet, that waste will vary by season, weekday and guest mix (Juvan et al., 2018). Enough baseline data is needed to understand the impact of those intervening variables before the behaviour change interventions are deployed in the field (Viglia and Dolnicar, 2020). At this stage (Time 1), we research designs can be pre-registered to increase the credibility of the research (Simmons et al., 2021).

When there is a substantial amount of variation in the dependent variable collected during the baseline measurement period, the behaviour change intervention is deployed and data collection continues. How long data needs to be collected depends on how frequently measurements can be taken. If only one measurement is taken per day (e.g. food waste generated per day), data collection will take longer to achieve the sample size required for statistical testing. As soon as sufficient data points are available for the pre- and post-intervention measurement, statistical tests can be run to determine if there is a significant difference and, if so, if the direction of the difference is as intended. The improvements achieved through the intervention (e.g. food waste generated per day) serve as evidence of social impact. In most instances, testing will need to account for all intervening variables. Table 1 lists examples of field studies testing behaviour change interventions in the context of climate change mitigation.

While Figure 2 looks like a simple experimental design, it is important to note that it is rarely possible in the field to conduct true field experiments because it is very difficult to ensure the random assignment of study participants to experimental conditions (Viglia and Dolnicar, 2020). For example, in hotel studies aimed at enticing tourists to behave in more environmentally sustainable ways, the researcher would have to randomly allocate when which tourists come to stay at the hotel. Because this is impossible, field studies are typically quasi-experimental in nature – they comply with all requirements for experiments except random allocation of people to experimental conditions. Note also that typically a survey experiment is conducted before a field study. The survey experiment does not aim to assess impact, but rather to assess whether the underlying theoretical mechanism works (see Zinn et al., 2024a).

Although in some instances practical recommendations can only be derived from theory, as is the case in space travel, in business research this is rarely the case because different dependent variables lead to different results (Viglia et al., 2024). For example, in a study aiming to develop an effective behaviour change intervention to reduce electricity consumption from air conditioning in a hotel, results from a survey, a virtual reality and a field study led to entirely different conclusions. Of the five interventions tested, the survey suggested none would reduce energy use, the virtual reality study concluded that interventions 3 and 5 would work and the field study determined that electricity use dropped for interventions 1, 2, 3 and 4 (Greene et al., 2024a). Similarly, in a study aiming to reduce food waste at all-you-can-eat buffets, the survey experiment suggested that some of the messages tested held substantial promise to change food waste behaviour, but when tested in the field none of these intentions translated into action and the level of food waste remained unchanged (Juvan et al., 2023).

Measuring actual impact is typically – but not always (e.g. online clicking data when booking or searching for information) – more labour-intensive and expensive than running a survey study or a survey experiment. Yet several studies have taken this approach to assess impacts across a wide range of behaviours (see Table 1), including: reducing household energy consumption via gamification where energy consumption was measured using household energy metres (Mulcahy et al., 2020, 2021); improving organisation-level sustainable waste behaviour where the number of waste bags disposed in the bin served as impact criterion (Tudor et al., 2008); reducing waste generated and increasing recycling and reducing electricity use in a construction company measured by waste and energy consumption per production unit (Jones et al., 2012); and organisational paper and cup recycling measured manually (Holland et al., 2006). Environmentally sustainable consumer behaviour is frequently investigated in the household context, using residential water consumption captured from water metres manually (Schultz et al, 2016); household electricity consumption extracted from electricity bills (Sapci and Considine, 2014); food waste generated by households obtained by manually weighing the waste (Li et al., 2021); public transportation and purchasing sustainable products (Bamberg, 2002); and household recycling (Alpizar and Gsottbauer, 2015).

In the vacation context, actual behaviour has also been used as a dependent variable, either to quantify the extent of an environmental challenge or to test the effectiveness of a behaviour change intervention. Behaviours studied in tourism include: booking high-emissions (Araña and León, 2016); using a fresh towel every day at a hotel (Goldstein et al., 2008; Baca-Motes et al., 2013); leaving uneaten food behind on the plate at a buffet (Kallbekken and Sælen, 2013; Dolnicar et al., 2020; Dolnicar et al., 2023a, 2023b; Zhu et al., 2024); requesting daily hotel room cleans (Dolnicar et al., 2019; Knezevic Cvelbar et al., 2021); electricity consumption (Dolnicar et al., 2017) and water use/shower duration in hotel rooms (Gössling et al., 2012); Tiefenbeck et al., 2019; Pereira-Doel et al., 2024); and food preparation waste in kitchens (Chawla et al., 2020).

Measuring real behaviour is necessary to quantify the problem, identify drivers of behaviour and evaluate longitudinal impact. Measuring buffet plate waste hourly, for example, reveals patterns of food waste generation across guest cohorts; early diners waste less food than diners during the last hour of operation (Juvan et al., 2021). Combining food waste and guest check-in data helps identify tourist segments that most contribute to plate waste (Juvan et al., 2018). Such descriptive insights inform the design of behaviour change interventions, the effectiveness of which is tested by re-measuring behaviour. Using buffet food waste as an example, a two-month field experiment demonstrated that placing a culture-specific message on dining tables reduced plate waste by 7% (Zhu et al., 2024). This insight is only possible when plate waste per person per day is measured before and after the table sign is deployed. Imposing monetary penalties for leaving food waste behind has proven less effective than positive incentives (Chang, 2022). Within the tourist segment of families, using a stamp collection game has proven to be highly effective – this intervention reduces plate waste among families dining at a hotel dinner buffet by 34% (Dolnicar et al., 2020).

It is not uncommon for findings from Mode 2 research to contradict those from Mode 1 research. For example, value-belief-norm theory routinely informs the design of behaviour change interventions, predominantly by communicating the negative environmental impact of certain actions (Stern et al., 1999). In contrast to value-belief-norm theory predictions, pro-environmental messaging does not always reduce buffet plate waste (Juvan et al., 2023). This does not mean that Mode 1 findings are invalid; rather, this result highlights the complexity of real-life environments and underscores the importance of evaluating the social impact of business research using objective behavioural measurements. An additional consideration in impact-oriented business research are unintended negative side-effects such as moral licencing (Barkemeyer et al., 2023) and negative emotions (Acuti et al., 2022). The potential of such unintended consequences can be assessed in manipulation checks (Zinn et al., 2024b), enabling industry collaborators to implement only interventions with a low risk of consumer backlash.

Business researchers do not routinely use actual behaviour in their studies because it is hard to measure. Manual measurement has the advantage that it is easy to implement and requires relatively low upfront costs, making manual measurement an effective option for single case studies. However, manual measurement is time-consuming and often requires substantial staff hours to collect the data (e.g. Chang, 2022), making manual measurement an unsustainable option for long-term projects or for studies that aim to collect behaviour data from several field sites. Despite the high external validity of the results (Viglia and Dolnicar, 2020), conclusions derived from single-site studies are limited to a specific consumer segment or business type. Most manual measurement-based studies also do not account for seasonal factors, which is an important aspect in business environmental performance (Sáez-Fernández et al., 2020).

An automatic measurement system can be particularly useful in countries with high labour costs, such as Australia. Despite the initial investment, such systems can support long-term behavioural measurement and deliver a high return on investment, making it suitable for longitudinal and multi-site studies. This aligns well with the definition of the social impact of research, which focuses on long-term benefits (Ruegg and Feller, 2003). Automatic measurement system also has high accuracy for quantifiable behaviours, such as water usage, electricity consumption and food waste. In the context of buffet food waste, for example, automatic measurement systems can capture waste and automatic people counters can count the number of diners (Zhu et al., 2024). This constant data feed allows researchers to monitor food waste in real time and provides a reliable dependent variable (food waste per person per day) to test the effectiveness of behavioural interventions (Dolnicar et al., 2023a, 2023b). The ongoing data flow from automatic measurement system can help researchers evaluate secondary outcomes of the research design, such as determining whether the intervention generates positive follow-up effects (e.g. behavioural consistency) or unintended negative consequences (e.g. moral licencing) in the long run.

During the United Nations 2030 Agenda meeting, the global community explicitly expressed that environmental sustainability issues can only be solved through collaboration between different stakeholders (United Nations, 2015). Research partnerships between academia and industry have led to significant outcomes, such as technological innovation and increased product competitiveness (Hall et al., 2001; Bikard et al., 2019). Such collaborations are equally critical for achieving social impact through business research. Researchers are multipliers in such collaborative partnerships; they serve as social “change agents” (Stephens et al., 2008) because of the diverse foci of academic research. From sustainable product development to policy recommendations, academic researchers create knowledge on how to improve environmental sustainability, making them key stakeholders in this discussion (Di Maria et al., 2019; Sjöö and Hellström, 2019). Industry relies on this knowledge to achieve sustainability goals (De Marchi, 2012), which positively affect firm performance (Di Maria et al., 2019). The diverse focus of academic research offers industry a range of potential behavioural interventions, allowing businesses to choose the low-hanging fruit – the most effective and practical changes – based on the nature of their operations. This creates a natural connection for academic-industry collaboration in the area of environmental innovation (De Marchi and Grandinetti, 2013), which pushes the research focus from theory to practical implementation. In such collaborations, researchers and industry partners often engage in a co-creation process, jointly developing research topics and designs that align with both practical needs and scientific knowledge advancement.

Policymakers also play an important role in facilitating knowledge transfer between academia and industry. Industry adopts newly developed sustainability approaches at a slow rate, largely due to the profit-driven nature of most organisations (Alexander, 2007). Not all industry actors are willing to change existing practices solely for the sake of environmental sustainability. In such cases, the role of policymakers becomes vital. For example, although solar panels can offer long-term financial and environmental benefits to businesses, many owners remain reluctant to install solar systems due to high upfront costs and long payback periods. In this context, low-interest loans or government subsidies become major factors in increasing the uptake of such technologies. Another important role for policymakers is to monitor and address the unintended consequences of sustainability initiatives, such as greenwashing (European Parliament, 2024). Although working with policymakers can be challenging for many researchers due to issues of accessibility and the long timeframe for policy change, it remains important for business researchers to showcase their work for greater exposure by, for example, producing and share video abstracts via social media, attending university media days to communicate their research with boarder audiences.

Calls for researchers to collaborate beyond their disciplines are not new, with discussions around interdisciplinary collaboration beginning in the early 1970s (Apostel, 1972). More than half a century later, this concept remains the exception rather than the norm in academic research. One of the primary reasons is the ongoing trend towards field specialisation and disciplinary fragmentation within academic research (Becher and Trowler, 2001; Davis, 2019). For many contemporary researchers, specialising in a niche area allows them to focus on a highly specific topic within their discipline to achieve a deeper understanding (Becher and Trowler, 2001). While this approach advances technological development in highly specialised fields, researchers who dive too deep into niche areas often become less aware of how their knowledge could be valuable to other disciplines. This creates research boundaries, which in turn lead to disciplinary fragmentation (Balietti et al., 2015).

A key characteristic of disciplinary fragmentation is the lack of collaboration across fields, which becomes particularly problematic when addressing “grand challenges” such as environmental sustainability and climate change (Kaldewey, 2018). To maximise impact in those complex issues, researchers need to look beyond their own fields and promote initiatives that leverage expertise from other areas of inquiry. For example, social scientists have substantial knowledge in behavioural change theories and can develop targeted behavioural change interventions based on these theories. However, much pro-environmental research within social sciences still relies on self-report methods (Greene et al., 2024b; Dolnicar et al., 2024b) or manual measurements to gauge behaviour (Chang, 2022), which either subject to social desirability bias (Zhu et al., 2024) or human measurement error.

The social impact of business research can be increased through collaborations with experts from other scholarly disciplines. For instance, Internet of Things (IoT) technology enables real-time monitoring of environmental indicators like energy consumption and water usage (Ullo and Sinha, 2020; Dolnicar et al., 2023a, 2023b). For social scientists, IoT offers an opportunity to obtain objective behavioural data, facilitating the shift from Mode 1 to Mode 2 research (Dolnicar et al., 2024a). This type of collaboration brings new perspectives to social science research because researchers no longer need to rely solely on self-reported data. Social scientists would be able to access large-scale, real-time field data to support their research, which provides more valid results and fuels new theory development. For engineers, such collaborations provide a valuable test ground to refine and adapt technology in real-world settings, potentially paving the way for future commercialisation. This reciprocal interdisciplinary collaboration is important for breaking down research boundaries and achieving maximum social impact.

Business research today increasingly focuses on generating social impact. However, many researchers struggle to achieve this goal due to a lack of clear guidance on how to translate theoretical knowledge (Mode 1 research) into practical, problem-solving mechanisms (Mode 2) that can deliver social impact. Our study addresses this gap by proposing a practical guide to help business researchers translate theoretical insights into measurable social impact.

We propose a framework that helps researchers design and implement impact-focused business research, which consists of five building blocks: leveraging theoretical and knowledge foundations, testing impact in field studies, measuring actual impact, collaborating with industry and policy makers and working across scholarly disciplines. Our framework advances the theoretical understanding of how social impact can be generated in business research by conceptualising the key factors as building blocks, which help business researchers identify critical elements contributing to translation for social impact.

Our framework provides a practical guideline for social impact generation. Table 2 serves as a checklist to guide researchers through all building blocks. Although the framework has been designed with climate change mitigation as the primary social impact, it is general enough to also be useful to other areas of social impact – such as economic, cultural or societal benefits.

The primary focus of our framework is to generate social impact by enhancing environmental sustainability. While from a systems perspective, environmental performance is the core component of social impact and can create spill-over effects to social and cultural domain, our framework is not specifically designed to target these two areas. Business researchers with a particular interest in the social and cultural domains may consider developing alternative frameworks to maximise social impact from these perspectives.