Success in the global marketplace can be achieved by developing and coordinating a firm's resources to address dynamic situations in the market. A firm can align its resources, including business models (BMs) more effectively, considering consumer requirements in the presence of dynamic capabilities. Such firms need to be flexible, to act on opportunities by making appropriate changes in culture and the organization. Different firms have different sets of dynamic capabilities. One firm may be good at innovation, while another may be good at identifying opportunities and exploiting them. Strong dynamic capabilities signify possession of better resources than competitors, which helps in gaining sustained competitive advantage. Firms with strong dynamic capabilities are able to create or modify resources, assets and capabilities, to address changing market conditions and customer demands. This helps in providing value to the customer over a longer time horizon. Setting up BMs during the initial phase depends equally on intuition and science (Teece, 2018). It is a part of dynamic capability which cannot be fully routinized (Teece, 2012), e.g., organizational processes. Dynamic capabilities cannot be easily replicated as they are built around history, routines, culture, and unique managerial characteristics (Teece, 2014). This uniqueness acts as a basis of sustained competitive advantage in firms.

Casadesus-Masanell & Ricart (2010, p. 100) observed that: “Strategy has been the primary building block of competitiveness over the past three decades, but in the future, the quest for sustainable advantage may well begin with the business model”. Having rare capabilities helps in creating competitiveness by linking BMs with strategy. BMs “…describe the design or architecture of the value creation, delivery, and capture mechanisms [a firm] employs. The essence of a business model is in defining the manner by which the enterprise delivers value to customers, entices customers to pay for value, and converts those payments to profit” (Teece, 2010, p. 172). A BM essentially identifies customers’ latent needs, the resources required to address those needs, and the capturing of value from important business functions (Teece, 2018). It needs to strike a balance between value creation, delivery and capture to be sustainable. BM elements need to be coherent, internally aligned (Ritter, 2014), and in line with the broader model for management used (Birkinshaw & Ansari, 2015). Good BMs are scalable across different market segments and help in differentiating firms from their competitors.

A company cannot be considered sustainable if it delivers an eco-friendly product, but uses a technology that harms the environment. The entire process needs to be sustainable, with a focus on surroundings and the entire ecosystem within which a firm operates. The term sustainable business models (SBMs) describes the process of organizations’ transformation to more sustainable economic systems (Rashid et al., 2013; Wells, 2013). SBMs are “business models that incorporate pro-active multi-stakeholder management, the creation of monetary and non-monetary value for a broad range of stakeholders, and hold a long-term perspective” (Geissdoerfer et al., 2018, p. 403). Today, such models are widely researched given their potential to provide competitive advantages (Porter & Kramer, 2011; Nidumolu et al., 2009). SBMs have a “global market perspective and take into account new industrialized countries” (Garetti & Taisch, 2012, p. 88). The definitions in extant literature portray SBMs as modifications to BMs, adding features and objectives. Changes can be made in terms of goals and/or value propositions, creation, delivery, and capture intended to create sustainability (Richardson, 2009). These are changes that can be made in the BM, to products and services, or at the base level (Bocken et al., 2014; Bocken & van Bogaert, 2016). SBMs emphasise stakeholder benefits and values, and not only customer/ shareholder value. Most researchers have defined SBMs as modifications to conventional BMs, with added propositions, value creation or value capture (Geissdoerfer et al., 2018). Bocken et al. (2014) categorized SBMs as the base of the pyramid, product-service systems, and a circular business model. Organizational design is a “critical managerial change lever” with the “dynamic capabilities of sensing, seizing, and transforming needed to execute SBMI process” (Minatogawa et.al., 2022, pp. 2–3).

The dynamic capabilities framework reflects the interdependence between BMs and strategy. Dynamic capabilities and strategy coalesce to form a sustainable BM, which steers organizational change, and facilitates a level of profits sufficient to maintain and improve organizational resources and capabilities. In firms with dynamic capabilities, BMs can be modified swiftly and effectively to meet changing market conditions. Dynamic capabilities play a significant role to improve organizational competitiveness by designing BMs linked to strategy. Dynamic capabilities coupled with effective strategies help formulate the BM, which transforms the organization (Teece, 2018).

Business model innovation (BMI) signifies innovation in creating, delivering and capturing value, thus attracting a customer base and enhancing profitability (Teece, 2010). According to Foss & Saebi (2017, p. 201), BMI reflects “designed, novel, and non-trivial changes to key elements of a firm's business model and/or the architecture linking these elements”. Technological innovations help corporations to be competitive (Şimşek et al., 2022; Perelygina et al., 2022; Wirtz et al., 2016; Zott et al., 2011) by providing novel products and channels of revenue (Chesbrough, 2010; Massa et al., 2017). Sometimes firms engage in BMI to meet environmental demands (Colovic, 2022). A BM needs to address customer needs profitably (Rumelt, 2012). The capacity of managers to enhance a BM has a fundamental role in dynamic capabilities (Teece, 2007). This signifies designing the initial model, reframing the model, and/or replacing the model over time. A successful BM begins with deep insights into the customer's needs and by looking into already existing models. Technology is an enabler for novel BMs. In highly developed economies, absolutely novel BMs are improbable, although not impossible. Most “new” BMs are similar to existing ones, and involve a combination of existing models or an improvement on existing ones. Choice of BMs depends upon how strong an organization's dynamic capabilities are (Teece et al., 2016). For instance, organizations with strong dynamic capability do not hesitate to innovate their BMs, which involves radical changes in resources or activities.

Strategy deals with how a firm will compete (Teece, 2018). One factor leading firms to seek BMI is the emergence of new technology (Kraus et al. 2019; Teece et al. 2016). The concept of dynamic capabilities posits that strategic resources lose value with time, as competitors catch up with them (Teece, 2018). Firms that are able to modify their capabilities better than their competitors are able to gain a sustained competitive advantage (Teece, 2007).

This study follows the scale development procedure suggested by Hinkin (1995), as it is deemed one of the most reliable and widely used scale development processes. It consists of five steps: (1) item generation; (2) content validity; (3) exploratory factor validation; (4) psychometric property assessment; and (5) nomological validity. Items developed were mapped with similar constructs from literature, qualitative insights and a combination of inductive and deductive methodologies were employed (Tanwar & Prasad, 2017). 20 in-depth interviews were conducted. The respondents were SME owners (12), and faculty members (8). Interviews lasted for 40–60 min on average, and revolved around issues in the implementation of different types of SBMIs across SMEs in Saudi Arabia.

Next, the items generated were structured into a questionnaire for data collection. First, we ran exploratory factor analysis (EFA), to test the factor structure and validate the scale. According to Marsh et al. (2014), EFA is less restrictive than confirmatory factor analysis (CFA), allowing ite to load onto respective factors, and cross-load with others, which ensures the robustness of a scale. As per the scale development procedure, EFA was followed by CFA and nomological validity assessment on a second dataset of 200 SMEs (Hinkin, 1995). CFA was conducted using Smart PLS 3 (following Ringle et al., 2005). PLS has become a widely used method to estimate path coefficients with a limited sample size and non-normality over the last decade (Hair et al., 2011). Finally, to check for nomological validity, items for competitive advantage and SME performance were added into the questionnaire. Fig. 1 summarizes the methodology applied in the study.

Fig. 1.

Scale development procedure.

Step 1: Item Generation

To generate items for the scale we interviewed 20 respondents, comprised of 14 males and 6 females. Of the 20 respondents, eight were PhDs, four were postgraduates, six were graduates and two were diploma holders. 10 were aged between 45–50 years, six were 40–44 years, and four were aged between 35–39 years. SBMI was conceptualized as a three-dimensional construct based on the works of Shakeel et al. (2020). The in-depth interviews generated 12 items: three for sustainable value proportion innovation, five for sustainable value creation/delivery innovation, and four for sustainable value capture innovation. The items in each construct were limited in number, as past research has concluded that increased item numbers reduce respondent validity (Bednar & Westphal, 2006). To enhance face validity, items were mapped to pre-existing works as shown in Table 2.

Step 2: Content Validity

Content validity describes how far measures represent a construct (Haynes et al., 1995). The items shown in Table 2 were checked by an independent panel, consisting of ten experts in the field, to assess face and content validity. Of these 10 experts, five were academicians, and five were SME owners. Taking a cue from Netemeyer et al. (1996), subject specialists were supplied with definitions of each construct. Each item was evaluated against three levels: ‘completely representative’, ‘somewhat representative’, and ‘not representative’ (Lin & Hsieh, 2011). Each item's content validity ratio (CVR) was in a range of between .75 and 1.00, with nine experts terming all the items essential for the scale. Therefore, the scale's content validity was established (Almanasreh et al., 2019; Bose et al., 2022).

Step 3: Exploratory Factor Analysis

Data were collected from SMEs located in Saudi Arabia for two main reasons. First, SMEs are regarded as the backbone of the whole Gulf region, as they account for nearly 60% of the gross domestic product and 80% of the workforce (PWC, 2016). The context of SMEs is important given the ambitious 2030 vision of the Kingdom, which aims to raise the contribution of SMEs towards GDP to 35% (from 20%). The vision further aims to focus on sustainable solutions for SMEs across the Kingdom of Saudi Arabia (Saudi Vision, 2030, 2016). Therefore, these two factors signify an appropriate research context. The sampling frame used was taken from the general authority of statistics of Saudi Arabia. To widen the scope of generalizability of the findings, a multi-industry sample design was adopted (as suggested by Katsikea et al., 2011). The sample for this study was taken in two stages. In the first stage, we followed a stratified sampling to identify manufacturing and service businesses that had made some sustainable innovations in their BMs in the past five years. Next, SMEs from each stratum were selected based on convenience sampling. We contacted the SME owners and managers by telephone between November and December 2021 to seek cooperation for participation in the study. The respondents were assured of anonymity of identity and confidentiality of their responses. The entire process was scrutinized and approved by the Ethics Committee of Qassim University.

400 questionnaires were distributed to SMEs across different industries. After a few reminders, 140 responses were received by the authors, 10 responses were discarded due to incompleteness, which resulted in a response rate of 32.5%. The final 130 responses were used for factor analysis. Table 3 highlights the demographics of the sample collected. Only those respondents involved in the strategic orientation of their company were recruited. The final sample was a combination of 34.6% senior managers and 65.3% middle management and was considered appropriate as observed in past BMI studies (Clauss, 2017). 67% of the SMEs were from the manufacturing sector and 32.3% were service based. 60% of the SMEs selected were functional for more than 10 years and 40% were in business for more than 20 years. 87% of the SMEs selected had local market orientation, 6.9% global and 5.4% regional. The ownership structure consisted of 91% private, 3% government and 6% mixed ownership (Table 3).

Data were tested for univariate normality. The skewness values ranged between −2 and +2, and kurtosis was between −7 and +7 for all constructs, implying univariate normality of the data (Hair et al., 1998; Yousaf et al., 2020). Next, EFA was conducted based on principal component analysis and varimax rotation, using SPSS 20. The Kaiser-Meyer-Olkin (KMO) value at .85 indicated that the sample was appropriate for factor analysis. Additionally, Bartlett's test of sphericity, based on the X2 value, showed the data's suitability for factor analysis. The findings revealed a 3-factor solution, with 65% variance being explained. However, taking cues from Field (2005) and Hair et al. (1998), items with a factor loading greater than .35 on other factors, and with communalities less than .50 were dropped. This resulted in the removal of two items (SVC5 and SRC4). EFA was run again on the revised list of items, and the findings for factor solution are highlighted in Table 4.

Step 4: Psychometric property evaluation

A second set of data was gathered from the same population used in Study 1. SMEs who had not participated in the first survey were used for data collection. Around 600 questionnaires were distributed, out of which 220 were received back. 20 responses were deleted due to missing information, and the remaining 200 were used for analysis, resulting in a 28.5% response rate. This sample contained 40% respondents from top management and 60% from middle management, focussing on managers responsible for setting their company's strategic orientation. The sample consisted of around 64.5% SMEs from the manufacturing industries, while 35.5% represented service organizations. 47% of the firms employed up to 50 people, 19% between 51–100, 15% between 101–150, 10% between 151–200 and 9% employed between 200–249. 65% of the SMEs were up to 10 years old, 18% were 11–20, 9% were 21–30, and 8% were 31–40 years old. 90% of the SMEs selected had a local market orientation, and 93% of SMEs had private ownership. Table 5 provides a snapshot of the sample demographics used for CFA.

Smart PLS 3 was used to check the factor structure in Table 4 and to verify the measurement model, following Ringle et al. (2015). Henseler et al. (2014) suggest that PLS is suitable in checking measurement model specifications. To derive standard error across each estimation within this measurement model, we used non-parametric bootstrapping with 5,000 samples (Hair et al., 2011). The scale was tested for indicator reliability, composite reliability, convergent validity, and discriminant validity. The factor loadings ranged from 0.718 to 0.854, significantly over the recommended indicator reliability cut-off of 0.4 (Bagozzi & Baumgartner, 1994). The composite reliability (CR) findings were between 0.846 and 0.911, which was considerably higher than the 0.6 cut-off value (Bagozzi et al., 1999). The average variance extracted (AVE) across each variable exceeded .50, and most factor loadings were greater than .70, indicating convergent validity (Fornell & Larcker, 1981). Cronbach's alpha, AVE and CR for each construct are shown in Table 6.

Furthermore, discriminant validity can be claimed, as the AVE's square root (Table 7) exceeds diagonal elements for the corresponding row in every construct (Fornell & Larcker, 1981). However, Henseler et al. (2015) found that Fornell & Larcker's (1981) discriminant validity assessment approach fails to uncover discriminant validity for structural equation models based on variance. They therefore put forward a different approach to assessment: heterotrait-monotrait (HTMT). HTMT is more robust, and has a 0.85 cut-off (Henseler et al., 2015; Rasoolimanesh et al., 2017). Values for all constructs were under this cut-off, implying evidence to support discriminant validity.

This study extends extant literature in three broad areas: i.e., SBMI, SME performance and competitive advantage. First, a comprehensive measurement scale for SBMI has been developed. Thereby, this study is one of the first attempts to contribute conceptually and methodologically to empirical SBMI research. Previous studies have conceptualized SBMI in a different manner; for instance, Upward & Jones (2016) provide an SBM ontology-based framework. However, their conceptualization fails to integrate sustainability and innovation. Meanwhile, França et al. (2017) created a strategic sustainable development framework. Bocken et al. (2014) developed archetype sets using value proportion, generation/delivery and capture, which are components of BMI. All these conceptualizations have significantly enhanced the body of SBMI research, but most are based on systematic reviews and case-based analysis. This study provided an opportunity to empirically validate SBMI. To develop a scale for SBMI, a rigorous methodological procedure was followed, which promises to aid future empirical studies in this area.

Second, the study has contributed to the extant SBMI and SME performance literature. Several studies have investigated the link between BMI and organizational performance (Bashir & Verma, 2019; Foss & Saebi, 2017; Dunford et al., 2010; Zott & Amit, 2007). To our knowledge, however, this study is one of the first that analyses the relationship between SBMI and performance in SMEs.

Third, the study contributes to the literature on SBMI and competitive advantage. In the current hypercompetitive environment, business outcomes are quite uncertain. Given the ever-evolving business environment, and changes in customer demands and preferences, firms need to be highly proactive in innovation. BMI has been established to assist organizations in achieving a competitive advantage (Peteraf & Barney, 2003; Amit & Zott, 2012; Bashir, Naqshbandi & Farooq, 2020; Naqshbandi & Kamel, 2017; Teece, 2010).

Fourth, the study is an attempt to explain SBMI using a dynamic capability perspective. The higher the dynamic capabilities in an organization, the greater their propensity to move towards an effective and sustainable form of business. Literature has established that SBMI results in a sustained competitive advantage (Amit & Zott, 2012; Bashir, Naqshbandi & Farooq, 2020). Thus, the study contributes to the existing theoretical base by explaining sustained competitive advantage and SBMI in SMEs through the dynamic capability approach.

The research findings presented should be viewed taking into consideration certain limitations. First, the results of this study are based on samples taken from SMEs in Saudi Arabia, which is culturally different from most of the world. Research suggests that there might be differences in BMI across cultures like China, which has higher organizational complexity (Clauss, 2017), or across other developed or developing nations. Therefore, future studies should try to validate the framework with a diverse sample from across the globe, to allow generalization of the results. Second, this scale is designed to measure general SBMI. Unique SBMI cannot be captured using this scale. Third, emphasis has been placed on the role of SBMI on SME performance and competitive advantage. However, there can be other factors which influence SBMI: events like the outbreak of COVID-19, industry lifecycle (Sabatier et al., 2012), further regulatory insights with respect to sustainability (Baden-Fuller & Haefliger, 2013; Drucker, 1984) and even competition within industries (Johnson et al., 2008). Future studies should seek to cover these factors.