Using entrepreneurship theory to inform owner-managed SME engagement with novel energy data and management tools: An action research study

SMEs and energy management: a persistent engagement challenge

SMEs are a heterogeneous and hard-to-characterise sector (Hampton & Fawcett, 2017). While energy strategy and large businesses have been studied (Cooremans, 2011), SMEs have not received the same attention (Janda et al 2014). Energy typically represents a small proportion of their overall operating costs, and this, combined with limited time and management capacity, means that energy management is rarely a priority (Revell & Blackburn 2007; Fawcett & Hampton 2020). Existing research has identified a wide range of barriers, including lack of awareness, limited access to capital, insufficient information and advice, and the complexity of the supplier and metering landscape (Trianni & Cagno 2012; Lunt et al. 2014).

At the root of many of these barriers is what has been termed an ‘attention deficit’: energy simply does not warrant sufficient attention to motivate action within the context of other business priorities (Revell & Blackburn 2007). This means that even where financial savings could be realised, they are often insufficient on their own to drive engagement. The literature suggests that framing energy management in ways that connect with broader business goals, such as operational productivity or quality, may be a more promising approach (Kenington et al. 2020).

Smart meter data-based interventions have attracted growing interest as a mechanism for improving energy visibility in SMEs. However, existing evidence suggests that presenting raw consumption data is insufficient to generate meaningful engagement without clear relevance to business activities (Fischer 2008). This underscores the need for tools designed with a detailed understanding of how SME owner-managers think and make decisions.

Entrepreneurship in SMEs: causation, effectuation and absence of strategy

SME business management research offers a range of theoretical constructs on how owner-managers make decisions. This paper focuses on Sarasvathy’s (2001) foundational work, which describes two contrasting decision-making logics. Causation is a goal-oriented approach in which a desired outcome is specified in advance and resources are deployed to achieve it. Effectuation, by contrast, is a means-based approach in which entrepreneurs work outwards from available resources, skills and networks to identify and exploit achievable goals. Rather than beginning with a fixed endpoint, effectual entrepreneurs ask ‘given what I have, what can I do?’ This mode of thinking tends to dominate in contexts of novelty and uncertainty, such as when developing new products or responding to new external opportunities.

Building on this, Hauser et al. (2020) introduced a third concept: Absence of Strategy. This describes situations in which neither Causation nor Effectuation is effectively deployed, manifesting as reactive, uncoordinated or neglected decision-making. This often occurs where activities lack sufficient priority or where the entrepreneur lacks the agency or resources to act.

Empirical studies show that SME owner-managers switch between heuristics depending on context (Smolka et al. 2018; Vershinina et al. 2017). Causation tends to dominate when managing established activities; Effectuation when exploring new opportunities; and Absence of Strategy when facing issues that fall below the threshold of meaningful attention. Crucially, these heuristics are sensitive to the specific decision context rather than being stable personality traits.

An unaddressed gap: connecting entrepreneurship and energy management research

Despite the richness of both the SME energy management literature and the SME entrepreneurship literature, these two fields have not previously been meaningfully connected. This is a significant gap. Cooremans (2011) previously explored the importance of strategic relevance of capital investments in energy efficiency, but this focused on large commercial organisations. If energy management decisions are shaped by the same entrepreneurial heuristics that govern many other SME decisions, then understanding those heuristics is likely to be helpful for designing effective interventions. The energy management implications of this are explored in the findings and discussion sections below.

Action research approach and rationale

This study employed an action research (AR) methodology, situated within the Participatory Research Paradigm (PRP) as described by Heron & Reason (1997). AR was chosen because the research question required direct engagement with the social and operational contexts in which energy-related decisions are made, rather than observation from a distance. AR operates on a participative epistemology in which knowledge is built through active engagement with participants across iterative cycles of diagnosing, planning, acting and evaluating (Coghlan & Brannick 2014).

AR operates simultaneously at two levels: a contextual level, generating practical knowledge relevant to the study context; and a meta-level, contributing to broader academic and methodological understanding. This dual-level inquiry is reflected in the research question and objectives.

The primary research question guiding the study was: How do owner-managed SMEs respond to the development and testing of new smart meter data-based energy management tools?

This was supported by four research objectives, operating at two levels of inquiry. At the contextual level:

• RO1: To inform the design and deployment of prototype energy management tools.

• RO2: To improve understanding of how energy management can be improved to reduce energy demand.

• RO3: To inform understanding about what other benefits smart meter data-based energy management tools may deliver, or whether energy management might be better considered as a co-benefit of other, higher priority actions.

At the meta-level of inquiry:

• RO4: To contribute to understanding about applying an action research approach in this context, learning about its strengths and weaknesses and developing recommendations for future work.

AR proved especially valuable in the context of SME energy management because energy-related decisions are deeply embedded in daily operational routines that would be difficult to surface through conventional qualitative methods. The participative nature of AR gave participants agency in shaping the direction of the research, which proved important for both the quality of the findings and the sustained engagement of participants.

Study design and context

The study was conducted in two interlocking parts. In Part 1, the researcher worked directly with Hildebrand Technology Ltd, a technology firm that participated in NDSEMIC and developed a prototype smart meter data-based energy management tool called GlowPro. This part involved facilitated group discussions, semi-structured interviews with potential tool users (shopping centre managers, retailers, hospitality businesses, security contractors and energy consultants), and iterative reflection to inform the tool’s design.

In Part 2, the researcher worked independently with three owner-managed SMEs over 18 months to test the GlowPro prototype and explore broader potential uses for smart meter and other operational data. The three businesses were: a small chain of cycle-shops (three premises in inner South London); a coffee-shop chain comprising two London sites and one site in South West England; and a coffee roastery in South West England. Together, the coffee businesses formed a group under a shared holding company structure.

Data collection in Part 2 comprised: in-depth initial interviews with owner-managers to understand business context; on-site visits to install GlowPro monitoring equipment and observe operational activities; semi-structured interviews to explore reactions to the GlowPro prototype; and follow-up interviews to explore potential broader uses for smart meter and other data. All participants were anonymised, with pseudonyms used throughout. Analysis drew on both inductive (grounded theory) and deductive (Hauser et al. 2020 framework) approaches, supported by NVivo for coding.

Analysis proceeded in two related stages. The first was a ‘within-cycle’ analysis in which the researcher wrote summary notes after each interview,¹ site visit and group discussion and, where relevant, shared interim interpretations back with participants in subsequent cycles to test and refine them. This dialogic checking is intrinsic to action research and provided a continual cross-validation of emerging interpretations (Reason 2006; Reason & Torbert 2001). The second stage was a retrospective analysis of the full dataset, including interview transcripts, field notes, observational records and supporting documents, coded in NVivo. Coding began inductively, following grounded-theory principles, to allow themes to emerge from the data itself. This inductive phase surfaced the relevance of entrepreneurial decision-making patterns and led to the identification of the Hauser et al. (2020) framework. A second, deductive process then re-coded the data using Causation, Effectuation and Absence of Strategy as analytic categories. Findings were derived where patterns were both consistent across participants and corroborated through participant feedback in subsequent inquiry cycles. The broad areas of unanticipated use reported in the findings, for example, emerged from this dual coding process and were confirmed in follow-up discussions with participants.

Part 1: Hildebrand and the development of GlowPro

Hildebrand’s response to the NDSEMIC Competition illustrates Effectuation in practice. Rather than beginning with a fixed tool design derived from energy management goals, the team began with their available means: existing energy monitoring hardware, relationships with shopping centres and the hospitality sector, and expertise in smart meter data. From these, they explored what new and useful products they could create.

In doing so, they identified that the core barrier to SME energy management was the attention deficit problem. This represented a deliberate strategic departure from the Competition’s explicit energy management goals, which the team judged to be necessary to produce something of genuine value to businesses.

Applying the Hauser et al. (2020) framework to these activities, Effectuation was the dominant mode throughout the tool development phase, with elements of Causation emerging as specific product features were developed and tested. Some Absence of Strategy was also observed, particularly in areas where the team encountered technical barriers or market complexities beyond their direct control, such as delays in the non-domestic smart meter roll-out.

Part 2: How owner-managed SMEs responded to GlowPro and smart meter data

Applying the Hauser framework to the findings

Applying Hauser et al.’s (2020) framework to the Part 2 findings reveals a consistent and meaningful pattern. When SME owner-managers engaged with smart meter and other data, the dominant cognitive mode was Effectuation: they explored what could be done with the data, given their existing operational priorities and resources, rather than asking whether the data could help them reduce energy. This Effectual process led to the identification of a range of novel potential uses for the data, which were not anticipated at the outset of the research.

Notably, however, the ultimate goals that participants identified for smart meter and other data were predominantly Causal in character: they wanted to gain greater control over existing processes and operations, and to manage their businesses more tightly and predictably. The Effectual exploration of possibilities was, in this sense, in service of a Causal end goal: enabling better Causation in areas of existing operations where they felt they lacked adequate control or visibility.

Energy management, by contrast, remained firmly in the territory of Absence of Strategy throughout. Despite showing Effectual creativity in identifying novel uses for smart meter data, participants did not spontaneously identify energy demand reduction as a goal. Energy management was not actively pursued because it did not offer sufficient control or productivity benefits in the form participants could relate to.

Energy management as absence of strategy: intervention design implications

The finding that energy management consistently falls within the Absence of Strategy heuristic is significant (although perhaps not surprising). This has important implications for intervention design.

Despite recent energy cost crises, for SMEs, energy bills typically represent a small proportion of operating costs for the kinds of SMEs studied here, and the cost savings from improved energy management are insufficient, in themselves, to motivate the sustained management attention that would be required (Revell & Blackburn 2007; O’Regan et al. 2005).

A more promising approach, suggested by this research, is to connect energy data with areas of the business where Causal decision-making already operates. The findings indicate that there appears to be genuine appetite for data tools that support those goals. If smart meter and other data can be made relevant to those existing priorities, energy management can emerge as a co-benefit rather than a primary goal.

Leveraging effectuation in engagement and intervention design

The Effectual quality of participants’ responses when exploring smart meter and other data suggests that giving participants the agency to interpret data within their own operational contexts is a productive engagement strategy. This aligns with broader findings in the AR literature and with the participative approach taken in this study (Heron & Reason 1997; Bradbury & Reason 2008).

Both Hildebrand and the SME participants demonstrated that when given the opportunity to explore smart meter data from the perspective of their own operational contexts, they consistently identified novel and valuable use cases that energy management researchers and policymakers would not have anticipated. This has important implications for how future innovation competitions, research projects and other policy interventions are designed: rather than specifying the desired tool or outcome in detail, leaving room for entrepreneurial exploration may generate more useful and widely adopted solutions.

Integrating smart meter data into existing systems

A consistent theme across the findings was the preference for integrating smart meter and other data into existing management systems, rather than deploying a new proprietary energy management tool. Participants identified POS systems and accounting platforms as natural integration points that they already used actively to manage their businesses. Integrating smart meter data into these systems would bring several benefits: it would make energy data visible alongside operational data; it would facilitate the creation of operational-level energy metrics meaningful to each business; and it would reduce the ‘attention cost’ of energy management by making it part of existing workflows.

This finding has significant policy implications. The evidence from this research suggests that targeting existing management systems providers, and encouraging them to integrate smart meter data into their platforms, may be a more effective route to improving SME energy management at scale. This represents a meaningful reframing of where innovation effort could helpfully be directed in future.

Energy productivity as a reframing for policy and practice

The findings also suggest a productive reframing for how energy management is communicated to SMEs. The dominant policy framings of energy efficiency and demand reduction are conceptually problematic for owner-managed SMEs who are primarily focused on growth and operational productivity. A framing around energy productivity – how efficiently (or perhaps better, effectively) energy is used in relation to operational outputs – aligns much more closely with how successful SME owner-managers think about their businesses.

Metrics such as energy used per cup of coffee served, or per kilogram of roasted coffee, are not only more meaningful to business owners but are also more likely to surface genuinely productive interventions – reducing energy waste in ways that also improve operational efficiency. Importantly, such an approach also offers a natural pathway toward engagement between energy policy and broader economic policy.

Study limitations

These findings draw on in-depth engagement with three owner-managed SMEs across two related sectors, alongside collaborative work with one technology partner. The intent was depth rather than breadth, consistent with the participative and qualitative nature of action research, in which transferability rather than statistical generalisation is the appropriate test of value (Heron & Reason 1997). The small number of cases means the specific patterns identified, particularly the four areas of unanticipated use for smart meter and other data, should be treated as illustrative rather than exhaustive. The broader theoretical contribution, applying the Causation, Effectuation and Absence of Strategy framework to SME energy management, rests on a logic of analytic resonance with the wider SME entrepreneurship literature (Sarasvathy, 2001; Smolka et al., 2018; Hauser et al., 2020). Further empirical work across a wider range of sectors, business sizes and national contexts would help test how far the patterns observed here apply more generally.