1 Introduction

In the course of digitalization, the strategic orientation of a small and medium-sized enterprise (SME) toward digital transformation is a decisive factor for long-term performance. While digitalization refers to utilizing digital technologies and digitized data (Brennen & Kreiss, 2016), digital transformation is broader and encompasses elements like digital orientation, strategy, and technology adoption (Kindermann et al., 2021; Vial, 2019). Hence, it can be understood as the various strategic postures concerning technology adoption and the resulting organizational changes (Yoo et al., 2012). This process is increasingly recognized as a key driver of competitive advantage across industries (Añón Higón & Bonvin, 2023; Nambisan et al., 2017; Schiuma et al., 2022), as digital transformation can lead to increased performance, efficiency, customer satisfaction, and loyalty (Autio et al., 2018; Eller et al., 2020; Rachinger et al., 2019). While larger corporations may approach digital transformation with extensive structural overhauls and significant investments, SMEs typically adopt more agile and resource-efficient approaches (OECD, 2021). Recently, the investigation of the subordinate constructs of digital transformation gained interest in research. In this regard, the construct of digital orientation emerged as a nascent strategic orientation toward digital transformation. This novel strategic orientation, rooted in the strategic alignment model (Henderson & Venkatraman, 1999), represents the extent to which organizations align their organizational processes and structures with the integration and usage of digital technologies (Kindermann et al., 2021; Leonardi, 2011).

However, the role of digital orientation in influencing organizational performance remains unclear. Studies have shown mixed results, with some finding no relationship at all and others indicating a positive effect (Kindermann et al., 2021; Nasiri et al., 2022). This ambiguity is particularly pronounced in SMEs, where, from a resource-based-view (RBV), limited resources and the need for agility create a different dynamic in comparison to larger corporations (Barney, 1991; Hess et al., 2016; Mithas et al., 2012). Despite their critical role in the economy, SMEs have been underrepresented in recent research exploring this relationship (Li et al., 2018; Verhoef et al., 2021; Vial, 2019). Notably, the unique characteristics of SMEs, such as their leaner organizational structures and more adaptable nature, characterized by their dynamic capabilities (DCs) (Teece, 2007), contrast with the more rigid and hierarchical nature of larger corporations. Consequently, a separate investigation into SMEs’ digital orientation and its impact on performance is imperative (Hoogendoorn et al., 2015).

So far, studies show that within the process of digital transformation SMEs appear to be less advanced compared to their larger counterparts (Eller et al., 2020). Considering the continuous development of technology, the identified lag of SMEs within the digital transformation process poses a potential threat, given the role of SMEs in national and global economies (Calderon-Monge & Ribeiro-Soriano, 2023). Despite the potential challenges, digital orientation as both a resource and a capability, can offer significant benefits for SMEs in the process of digital transformation. Subsequently, digital technologies can streamline operations, improve customer service, and open up new market opportunities, especially relevant for SMEs due to their typically closer customer relationships, more nimble operations, and lack of resources (Barney, 1991; Bharadwaj et al., 2013; Felicetti et al., 2024; Mithas et al., 2012). Additionally, the smaller scale of SMEs often allows for quicker implementation of digital solutions, although this is countered by a more pronounced sensitivity to the initial costs and resource requirements of such transformations (Aldrich & Auster, 1986; Hess et al., 2016; Ulas, 2019; Vial, 2019). Therefore, understanding the relationship between digital orientation and SME performance, and identifying strategies to support SMEs in their digital transformation journey, is of utmost importance (Dörr et al., 2023; Felicetti et al., 2024; Nasiri et al., 2022; Vial, 2019).

Consequently, the aim of this paper is to shed light on the relationship between digital orientation and SME performance, and to answer the research question of whether digital orientation has an impact on SME performance. To adequately answer this question, we collected a dataset from 1135 SMEs, which contains both quantitative (financial information) and qualitative (website-text data) information, avoiding common data bias through the use of data triangulation (Jick, 1979). Our results show that the impact of digital orientation and certain subdimensions on SME performance follows a U-shaped relationship. This implies that the initial phase of digital orientation is accompanied by a decline in performance. In the subsequent phase, the trough of performance decline is reached, and the performance benefits emerge. Consequently, this study contributes to the explanation of why some SMEs avoided or stopped the strategic initialization of the digital transformation process considering the initially negative effects on SME performance. More specifically, this study indicates that SMEs should focus on the long-term benefits of digital orientation, as initial challenges are eventually offset, resulting in enhanced SME performance over time.

2 Theoretical framework

2.1 Digital transformation and digital orientation in SMEs

Digital entrepreneurship can be defined as the exploitation of digital technologies in entrepreneurial processes and ventures (Berman et al., 2023; Davidson & Vaast, 2010; Kraus et al., 2019) and plays a critical role for SMEs in harnessing digital technologies to lower barriers to market entry and enable global competitiveness (Autio et al., 2018; Nambisan et al., 2017). Furthermore, digital entrepreneurship supports SMEs in developing adaptive strategies that leverage digital capabilities to exploit new market opportunities effectively (Autio et al., 2018; Berman et al., 2023; Kraus et al., 2023). Digital transformation as an operationalization of digital entrepreneurship can be defined as a process of organizational improvement via substantial change in the organization’s characteristics through the implementation and utilization of digital technologies (Vial, 2019). Accordingly, digital transformation is based on the integration and adaptation of information-, computer-, communication-, and connectivity-technologies, i.e., digital technologies, within an organization (Reis et al., 2018; Vial, 2019). The process of digital transformation is particularly pertinent to SMEs as it often revolves around integrating and adapting to digital technologies within limited resource environments, leading to distinct transformation pathways compared to larger corporations (Aldrich & Auster, 1986; Li et al., 2018; Ulas, 2019).

Consequently, SMEs are at the core of a controversial scientific debate about their ability to handle the changes implied by digital transformation. Central to the current discussion is the contested ability of SMEs to adopt and implement digital technologies (Eller et al., 2020; Nguyen et al., 2015), as SMEs are considered to be more risk averse, especially in terms of adopting digital technologies (Ahmad et al., 2014; Kallmuenzer et al., 2024; Taylor & Murphy, 2004).

A strategic approach to digital transformation is central to a successful organizational transition (Ghobakhloo & Fathi, 2019; Kraus et al., 2022). This often results in a competitive advantage which constitutes a valuable intangible capability that is difficult for others to imitate (Kindermann et al., 2021; Schweiger et al., 2019; Teece, 2007). The first step to implement a strategic approach to digital transformation is the creation of a digital orientation. Specifically, digital orientation is defined as: “[...] an organization’s guiding principle to pursue digital technology-enabled opportunities to achieve competitive advantage.” (Kindermann et al., 2021, p. 649). Digital orientation is rooted in the strategic alignment model (Henderson & Venkatraman, 1999) and describes the extent to which organizations adjust their business strategy to create value through digital alignment (Kindermann et al., 2021). Therefore, digital orientation can be understood as an organization’s fundamental perspective on the phenomenon from which the digital strategy is derived in order to successfully master the digital transformation (Bharadwaj et al., 2013; Kindermann et al., 2021; Vial, 2019). This strategic adaptation aligns closely with the concept of digital entrepreneurship, which emphasizes leveraging digital capabilities to foster innovative and competitive business practices in SMEs (Berman et al., 2023; Felicetti et al., 2024; Kraus et al., 2023).

Generally, the construct of digital orientation can be categorized into four subdimensions, each of which describes different facets. The first subdimension, “architecture configuration” involves the digital infrastructure and explains how organizations facilitate the digital transformation of products and processes (Kindermann et al., 2021). The second subdimension “capabilities” comprises human capacities, i.e., the skills and knowledge required in connection with an organization’s digital transformation (Kindermann et al., 2021). It includes both the application-specific characteristics and the management-related aspects necessary to implement such a process within the organization (Kindermann et al., 2021). The third subdimension “ecosystem coordination” primarily describes inter-organization and customer relations for the use of information access and network effects within the digital transformation process (Kindermann et al., 2021). The fourth and final subdimension, “technology scope” describes digital technologies, as well as competencies and functionalities that are specifically offered for the customer. The technology scope is characterized by a “[...] set of digital technologies that allow the organization to realize strategic growth.” (Kindermann et al., 2021, p. 648). Together, these four subdimensions capture the digital transformation-related value creation, the emergence of new skills, and the advancement of existing skills and organizational prerequisites induced by digital technologies.

2.2 Resource-based-view and dynamic capabilities in SMEs

The debate about digital transformation in SMEs reflects the essence of the RBV (Barney, 1991), which posits that an organization’s resource constraints significantly influence its competitive advantage. Additionally, these resource constraints affect SMEs in their digital transformation journey (Elia et al., 2021). In fact, SMEs face distinct challenges in implementing digital transformation compared to larger corporations, primarily due to three constraints: limited financial access, constrained human capital, and challenges in scaling business models (Li et al., 2018; Priyono et al., 2020; Jin Zhang et al., 2015). Consequently, SME’s frequently fail to grasp the opportunities associated with digital technologies (Nguyen et al., 2015). From an RBV perspective, this can be attributed to the liability of smallness (Aldrich & Auster, 1986), which states that the efficient implementation of resource-intensive digital technologies often requires resources that are limited (Clohessy & Acton, 2019; Meyer, 2011).

However, despite these constraints, SMEs can leverage their inherent agility and entrepreneurial spirit — key resources as per RBV — to create unique value in digital transformation (Vrontis et al., 2022). Thus, SMEs’ smallness can also be seen as one of their key advantages, because their size implies a certain degree of flexibility and thereby fosters agility and adaptability to new trends and changes in their environment (Levy & Powell, 1998). This is especially beneficial in the current digital era, where the ability to quickly adopt new technologies and processes can be seen as a major competitive advantage. Indeed, previous research has shown that SMEs generally respond very quickly to external changes and are highly adaptable when it comes to digital technologies (Aldrich & Auster, 1986; Zhu et al., 2006; Kraus et al., 2022). In line with this, DCs enable SMEs to continually adapt and reconfigure their organizational resources and strategies in response to rapidly changing digital environments (Felicetti et al., 2024; Kraus et al., 2022). Consequently, considering the DC approach as an additional theoretical lens offers deeper insights into how SMEs can effectively navigate digital transformation (Schneider et al., 2023).

DCs are an organization’s abilities to integrate, build, and reconfigure internal and external competencies to address rapidly changing environments (Eisenhardt & Martin, 2000). Specifically, this approach includes three core mechanisms: sensing new opportunities, seizing these opportunities through resource and skill development, and reconfiguring organizational assets to enhance competitiveness (Eisenhardt & Martin, 2000; Teece, 2007). This aligns with the notion that digital transformation requires SMEs to not just adapt to new technologies but to strategically use these competencies to redefine their business logic and processes (Gong & Ribiere, 2021; Vial, 2019). In this sense, digital transformation can be seen as a process that is not static but continuously evolving, requiring SMEs to develop higher-order capabilities that extend beyond routine operational capabilities (Felin & Powell, 2016; Winter, 2003). These capabilities, including process efficiency, are crucial for SMEs to derive performance benefits from digital transformation (Helfat, 2022).

These observations connect with both the RBV and DC frameworks, where the strategic allocation of resources toward digital strategies exemplifies the DCs necessary for successful digital transformation (Leso et al., 2023). Consequently, successful digital transformation in SMEs requires a combination of technological and management skills (Li et al., 2018; Zhang et al., 2022) which underlines the importance of considering both RBV and DC approaches when addressing digital transformation in SMEs.

In this context, digital orientation can play a crucial role in guiding SMEs’ digital transformation efforts. By providing a strategic framework for the adoption and use of digital technologies, digital orientation can help SMEs to navigate the complexities of the digital era, and to identify and seize the opportunities that digital technologies present by fostering their DCs (Quinton et al., 2018; Teece, 2007). However, achieving a successful digital orientation requires more than just the adoption of digital technologies. It also involves a fundamental shift in the organization’s strategic focus, toward a more integrated and holistic approach to digital transformation (Kindermann et al., 2021). This transition, albeit demanding for SMEs, remains an integral stride toward harnessing the full capacity of digital technologies to foster business enhancement and competitive advantages.

Although the scientific discourse on digital orientation has recently gained attention, SMEs have mostly been neglected in previous studies (Ardito et al., 2021; Rupeika-Apoga et al., 2022; Saunila et al., 2021). We observe this as a significant gap in the literature because the unique challenges and opportunities SMEs face in the process of digital transformation have critical implications. For instance, from a RBV perspective, SMEs often have resource constraints (Aldrich & Auster, 1986; Barney, 1991), which limit their capacity to fully engage with and exploit the benefits of digital technologies (Elia et al., 2021). This is especially concerning as SMEs represent the backbone of many economies worldwide (Ardito et al., 2021; Rupeika-Apoga et al., 2022). Specifically, SMEs contribute significantly to GDP (Ayyagari et al., 2011), employment (Beck et al., 2005), and innovation (Audretsch & Keilbach, 2007). They are responsible for a significant portion of job creation (Ayyagari et al., 2011; Beck et al., 2005), fostering entrepreneurial spirit and innovation, and often provide the foundation for the growth of new industries (Audretsch & Keilbach, 2007). Therefore, understanding the digital orientation of SMEs is not only important from an academic perspective, but also has significant practical implications in the digital era for policymakers and practitioners seeking to stimulate economic growth and competitiveness.

3 Hypotheses development

The potential rewards of establishing and pursuing a digital orientation originate from the integration of digital technologies into a concrete business strategy and culture (Kane et al., 2015). Hence, digital orientation is accompanied by a change in resources and capabilities, which decisively influence the future direction and therefore also the performance of an organization (Schiuma et al., 2022; Wang et al., 2020). The rise of digital orientation as an increasingly important strategic orientation (Bharadwaj et al., 2013) follows recent insights that digital orientation can be considered a valuable, non-imitable resource (Schweiger et al., 2019), i.e., a DC, that can contribute to an organization’s competitive advantage (Kindermann et al., 2021; Schiuma et al., 2022).

Nevertheless, previous research on the effect of digital orientation on organizational performance is inconclusive. Nasiri et al. (2022) found, that for large corporations pursuing a digital orientation does not have a significant effect on organizational performance. However, this argument is relativized by the fact that digital orientation has a positive impact on organizational performance if there is a certain minimum level of digital maturity in the corresponding organization (Nasiri et al., 2022). This implies that the initial costs of starting the digital transformation process include investments that are negatively correlated with organizational performance. The initial negative impact of digital orientation may act as a barrier for SMEs to initiate the process of digital transformation due to resource constraints which can be attributed to the RBV and more specifically the liability of smallness (Aldrich & Auster, 1986; OECD, 2017). As many SMEs start the digital transformation process from scratch, they need to heavily invest resources in new digital technologies and the development of required DCs, leading to short-term performance losses due to the high costs and potential disruptions associated with implementing these technologies (Besson & Rowe, 2012; Ulas, 2019). However, the negative effect reverses over time if the organization intensifies and proceeds with the digital transformation process (Nasiri et al., 2022). Accordingly, the initial resource investments can lead to long-term performance gains as the SMEs become more efficient and competitive due to their digital orientation (Nasiri et al., 2022). Accordingly, the impact of digital orientation on organizational performance is not linear but curvilinear, following a U-shaped pattern. This U-shaped relationship is the result of the dynamics involved in the process of digital transformation, which includes an initial resource investment phase. This resource investment phase, according to the RBV, is often associated with a decrease in performance, followed by a phase of growth and performance enhancement through DCs, which are in turn fostered by digital orientation. Therefore, we hypothesize:

  • \(H_1\)   There is a curvilinear relationship between digital orientation and SME performance that follows a U-shape.

Building upon the overarching construct of digital orientation, this study further delves into its subdimensions: architecture configuration, capabilities, ecosystem coordination, and technology scope. Each of these dimensions represents a unique aspect of an SME’s digital orientation and can independently influence the organization’s performance (Kindermann et al., 2021). Investigating these dimensions separately enables a more nuanced understanding of how different facets of digital orientation contribute to SME performance.

The first dimension considered is architecture configuration. The architecture configuration dimension of digital orientation refers to how an SME organizes its digital resources and capabilities. An effective architecture configuration can enable an SME to better leverage its digital resources, leading to improved performance (Kindermann et al., 2021). Nevertheless, the process of evolving and implementing an efficacious architecture configuration is not devoid of challenges. For instance, it has been established that the productivity benefits of digital technologies may be delayed due to the time required to reconfigure work processes and develop complementary skills (Dewan & Min, 1997). Consequently, a sufficient architecture configuration demands significant upfront resource investments in terms of capital, time, and human resources, which might initially lower organizational performance, resulting in a U-shaped relationship (Besson & Rowe, 2012; Ulas, 2019). Moreover, effective architecture configuration entails not only the technical aspects of integrating various digital technologies but also the alignment of these technologies with the business strategy, which may require organizational learning and adaptation (Sebastian et al., 2017) ultimately building a unique DC. The complexity and effort involved in these processes potentially explain the initial decrease in performance. However, once the configuration is successfully implemented and aligned with the business strategy, the DCs and the emerging benefits begin to outweigh the costs, leading to improved performance. Therefore, in light of the aforementioned insights, we hypothesize:

  • \(H_2\)   There is a curvilinear relationship between the dimension architecture configuration and SME performance that follows a U-shape.

The capabilities dimension of digital orientation refers to the skills and competencies that an SME has in using digital technologies. Developing these capabilities can enhance an SME’s ability to utilize digital technologies as a DC for its competitive advantage, leading to improved performance (Kindermann et al., 2021). The capability to configure and reconfigure digital technologies can provide a basis for a competitive advantage by allowing firms to respond rapidly and effectively to changes in the external environment (Teece, 2009). Furthermore, digital capabilities have been emphasized as instrumental in shaping the ability of SMEs to exploit digital technologies, to seize the strategic opportunities offered by digital technologies, and mitigate potential threats, thereby fostering superior performance (Kane et al., 2015; Li et al., 2018). Simultaneously, a high level of digital capabilities can also result in complexity and increased coordination costs, especially considering resources, particularly in the initial stages (Kane & Alavi, 2007). This complexity is due to the dynamic and rapidly changing nature of digital technologies, which necessitate continuous learning and adaptation. In this regard, the upskilling process can necessitate significant initial investment in training and development, potentially causing a decrease in available resources, and immediate performance and leading to a U-shaped relationship with performance (Besson & Rowe, 2012; Ulas, 2019). Consequently, while the development of digital capabilities is advantageous for SMEs in order to build DCs, it is imperative for them to manage this development carefully, especially from a RBV perspective considering the potential initial resource costs and risks. Therefore, we hypothesize:

  • \(H_3\)   There is a curvilinear relationship between the dimension capabilities and SME performance that follows a U-shape.

The ecosystem coordination dimension of digital orientation refers to how an SME coordinates its interactions with customers, suppliers, and other stakeholders, through digital technologies. Ecosystem coordination via digital channels can not only augment an SME’s ability to swiftly respond to market changes but also strengthen customer relationships and value creation within its supply chain, all of which can lead to DC and therefore improved performance (Cenamor et al., 2019; Nambisan et al., 2017). For example, the use of digital platforms enables firms to have a real-time connection with customers, capturing their feedback and preferences to better align products and services (Nambisan et al., 2017). Additionally, effective digital ecosystem coordination can facilitate knowledge sharing and collaboration with suppliers, enhancing the overall value chain (Vargo et al., 2011). Despite the potential advantages, the establishment of effective digital ecosystem coordination is not straightforward. It often requires significant upfront investments in digital platforms and systems (Besson & Rowe, 2012; Ulas, 2019). Furthermore, from a RBV perspective, the potential complexity and novelty of new digital ecosystems could require a learning and adaptation phase, leading to potential delays in the realization of performance improvements due to upfront investments of resources. These initial resource investments and adjustment periods could result in a U-shaped relationship with performance, where the benefits become noticeable only after a certain level of investment and utilization has been reached and DCs can be achieved. Hence, the following hypothesis is proposed:

  • \(H_4\)   There is a curvilinear relationship between the dimension ecosystem coordination and SME performance that follows a U-shape.

The technology scope dimension of digital orientation refers to the range and diversity of digital technologies that an SME leverages. Once the digital technologies are integrated and utilized effectively, the SME can begin to see the benefits of a broader technology scope. For example, Kindermann et al. (2021) found that a wider range of digital capabilities can lead to improved performance as they provide more opportunities for process automation, data analysis, and customer engagement. Likewise, digital technologies can enable SMEs to develop new products and services, improve their operational efficiency, and enhance their competitive position (Ulas, 2019). Hence, a broad technology scope can enable an SME to utilize a wide range of digital capabilities, leading to utilizing DCs and improved performance (Kindermann et al., 2021). From an RBV perspective, in the initial stages, however, the investments in acquiring and implementing a range of digital technologies can directly strain an SME’s available resources, leading to a decrease in performance. Accordingly, resource investments in digital technologies can initially lead to a decline in firm performance due to the costs and challenges of implementation as the organization learns to leverage these technologies efficiently. However, the resulting DCs could lead to performance benefits that are expected to improve over time, proposing a U-shaped curve (Besson & Rowe, 2012; Kindermann et al., 2021; Ulas, 2019). Therefore, we hypothesize that an SME’s performance might initially decline as it expands its technology scope, but as the SME learns to leverage these technologies effectively, its DC and therefore its performance will increase, resulting in a U-shaped relationship:

  • \(H_5\)   There is a curvilinear relationship between the dimension technology scope and SME performance that follows a U-shape.

Table 1 Sample creation and operationalization process
Full size table

4 Methodology

4.1 Sample and data

In order to create a holistic and accurate picture of the current status of digital orientation, it is favorable to gather as much information about an organization as possible. Accordingly, to generate a representative sample, we collected qualitative and quantitative information on our sample of SMEs (see Table 1). First, we created a list of SMEs and exported it from Amadeus (Bureau van Dijk, 2023), a database containing a wide range of information for public and private organizations across Europe. Based on this list, we created a sample of SMEs by having an annual turnover equal to or below 50 m. € and a staff headcount below 250 (European Union, 2003; German Presidency of the Council of the European Union, 2020). Further, only SMEs that had full available information on the following indicators: URL, profit (last three years), turnover (last three years), number of employees (last three years), export, number of trademarks, industry, and organization age, were selected. The SMEs of our final sample are from different European countries and industries. After the first selection process, our sample from the Amadeus database consists of 12,871 SMEs.

In the next step, we utilized web scraping techniques to capture the website texts for each of the URLs in the sample obtained from Amadeus. We initiated this process in late 2022. This allows us to collect qualitative data (text), including the text of up to the first 100 sub-pages of the considered SME’ websites. Accordingly, we generate a second dataset, next to the quantitative Amadeus data (financial), that contains the qualitative text data of the website from each considered SME. The obtained text data is diverse and includes everything an SME published on its website including company descriptions, mission statements, product descriptions, company news, and more. This bidimensional data base involving the combination of qualitative website-text data and quantitative data on SME performance, allows us to explore the relationship between an SME’s digital orientation and its performance in a comprehensive and nuanced manner.

Subsequently, we started the operationalization process for both the Amadeus dataset (financial data) and the dataset with scraped website text. The operationalization process differs for the qualitative (website-text) and the quantitative (Amadeus financial information) data of our final sample. For the text data, we started the operationalization by including only German- and English-language website-texts in our analysis, as we used both the original English language dictionary and a version translated into German. The proficiency of both authors in German and English enables the utilization of translated dictionaries to verify the results. We also removed missing observations and dropped subpages, which are not of interest to our analysis or could potentially bias the results (e.g., cookies and privacy). We then followed standard data processing techniques and removed all non-textual elements, symbols, and punctuation from the collected website texts and converted the text to lowercase to allow for case-sensitive comparability. In the next step, we tokenized the text of each website to perform analysis on a per-word basis. We also converted the website words into both lemmas and word stems for comparability and standardization. Lemmatization and stemming are standard procedures in text-based analysis, that enable the identification and comparability of words regardless of their grammatical structure (Balakrishnan & Ethel, 2014; McKenny et al., 2018). In order to measure digital orientation, we utilized the construct by Kindermann et al. (2021), which can be defined by a set of words called the digital orientation dictionary. We have translated this construct into German to analyze the text data from German websites adequately. In order to match the words from the dictionary with the lemmatized/stemmed website text, we also applied the same techniques (lemmatization, stemming) to all digital orientation words. After the cleaning process, which included dropping all observations where no text could be scraped and aggregating all website-related text, we obtained 1652 SMEs in our final qualitative text data.

For the quantitative data, we started the operationalization by dropping missing values and duplicates. Additionally, we controlled for all unrealistic or falsely documented values. In the last step, we match the quantitative Amadeus data of our sample SMEs with the website-text data gathered for each SME. After the merge 1,135 SMEs remain in our final sample including complete information on every organization’s website texts, profit (last three years), turnover (last three years), number of employees (last three years), export, number of trademarks, industry, and organization age.

4.2 Measures

The current state of research on extracting the level of digital transformation, especially digital orientation, is rather scarce and subject to intensive criticism (Thordsen et al., 2020). We therefore apply a novel analysis in this context: the computer-aided text analysis (CATA) (Krippendorf, 2004), building on an established and validated dictionary representing the construct of digital orientation (Kindermann et al., 2021; McKenny et al., 2013; Short et al., 2010). The dictionary consists of all relevant words that represent the phenomenon of digital orientation as a whole. Further, the dictionary can be subdivided into four subdomains (architecture configuration, capabilities, ecosystem coordination, and technology scope) where each sub dictionary is representative of the corresponding sub dimension.

For our study, we use the original English version of the validated digital orientation dictionary as well as a version translated to German, to also check websites that contain only German language. Previous research used text analysis mostly on a sample of annual reports (Boling et al., 2016) or letters to shareholders (Grühn et al., 2017; Kindermann et al., 2021; Short et al., 2010). As we, however, focus on SMEs, annual reports or letters to shareholders are mostly not available. In contrast, nowadays almost every organization has a website. We build on this fact and use the existing information of the website texts to derive our set of independent variables (Esrock & Leichty, 2000).

The derivation of our independent variable, the digital orientation score of an SME, is rooted in two theoretical assumptions: The Sapir-Whorf hypothesis on the one hand, which states that the frequency of occurrence of a word determines the direction and intensity of decision-makers’ attention to a given topic (Abrahamson & Hambrick, 1997). On the other hand, digital orientation behavior is a result of decision makers’ attention, so frequent usage of words related to strategic orientations is associated with more attention toward the phenomenon (Grühn et al., 2017; McKenny et al., 2018). Furthermore, in contrast to a concrete strategy, which may be hard to extract from a website unless it is not mentioned explicitly, the construct of strategic orientation, i.e., digital orientation, is a broader concept that reflects an organization’s overall perspective and approach or principle to managing its resources, processes, and activities (Hakala, 2011). Hence, an orientation resembles an organization’s general philosophy that guides its strategic decision-making (Morgan & Strong, 2003) which, at least partly, is inherent within the text provided on a website.

4.3 Dependent variable

To investigate organizational performance, we test the explanatory variables against a composed growth indicator, as previous research showed that growth is a suitable predictor of SME’ performance (Bhatti & Awan, 2014; Parmenter, 2010). We utilize this indicator to generate a differentiated picture of SME performance and to account for changes over time. As performance measurement systems in general should be designed to capture data over time, providing a more comprehensive view of an organization’s performance (Neely et al., 2005), we have chosen to examine the growth rates for the number of employees, turnover, and profit over the last three available years (2019–2021) (Coad & Höllzl, 2012). This period was selected as it provides on the one hand a more comprehensive picture of an organization’s performance than a snapshot (single-year data), and on the other hand, it is the most recent data we could obtain from the Amadeus database. It allows us to capture any latent effects of strategic shifts or investments in digital orientation that may not immediately manifest in the same year. While in many studies, total sales or the number of employees is used as an indicator of overall organizational performance, we use growth rates for the number of employees, turnover, and profit (Coad & Höllzl, 2012). Since a one-sided consideration of growth variables can lead to distortions, we combine several variables in order to obtain a composed growth indicator as a representation of organizational performance (Weinzimmer et al., 1998). To achieve normal distribution all three development variables are individually categorized into 10 identical categories (see Table 2). Following, we establish an average development rate of annual turnover (= turnover growth rate), annual profit (= profit growth rate), and annual employees (= employee growth rate) from the last three observed years as an indicator of turnover, profit, and employee development (Richard et al., 2009). By summing these three indicators and calculating the mean, we are able to construct an estimator of organizational performance that is in line with existing research on organizational performance measurement (Weinzimmer et al., 1998).

Table 2 Dependent variable categories
Full size table

4.4 Explanatory variables

To obtain our key independent variables for digital orientation, we followed the established four-dimensional operationalization of digital orientation. Accordingly, we create five independent variables, one for each of the subdimensions: architecture configuration, capabilities, ecosystem coordination, and technology scope, and one for the overall superordinate construct of digital orientation. In line with previous research, we apply multiple steps to create our independent variables for digital orientation and its subdimensions (Grühn et al., 2017; Kindermann et al., 2021). First, we count the number of words that match the respective subdimensions and dictionaries on each website-text, in order to obtain a score that reflects the overall digital orientation of an SME. We follow prior research by calculating one aggregated score for each variable (Short et al., 2010). Second, because of the varying length of text on the websites of our sample SMEs (see Table 3, Word Count), we normalize the score to the base of 1000 words to enable comparability. Finally, to minimize distortions, we control for outliers. We check the consistency of our digital orientation measure following established rules of former CATA research through the computation of correlations (Covin & Wales, 2019; Short et al., 2018). Descriptive statistics of the variables are shown in Table 3.

Table 3 Summary statistics
Full size table

With respect to prior research, we considered several other variables that may affect the relationship between digital orientation and organizational performance. We control for the industry sector with a binary indicator of the manufacturing sector. Also, we control for organizational size and age, as previous studies indicated that both are affecting organizational performance due to the liability of smallness and liability of newness (Aldrich & Auster, 1986; Hite & Hesterly, 2001). Also, we control for the number of trademarks as an indicator for innovation activity (Mendonça et al., 2004) and exports as an indicator of internationalization of an organization (Añón Higón & Bonvin, 2023). The description and information of the control and other variables are shown in Table 4.

Table 4 List of variables
Full size table

The final sample encompasses a diverse array of SMEs to provide a comprehensive overview of digital orientation across different sectors and geographical contexts. The geographical distribution of our sample, detailed in Table 5, demonstrates a significant representation from various European countries, with a notable concentration in Great Britain and France. This geographical spread ensures that our findings capture the varied digital orientation practices influenced by different national business environments. Furthermore, the age distribution of the organizations in our sample, presented in Table 6, ranges from newly established to well-established entities, providing insights into how organizational age might impact digital orientation and technology adoption. The majority of our sample SMEs are located in the category 20–30 years. Sector-wise, as shown in Table 7, our sample is heavily represented by the manufacturing sector, followed by wholesale and retail trade, and professional, scientific, and technical activities. This sectoral diversity allows us to examine digital orientation in SMEs across a broad spectrum of economic activities. Together, these demographic characteristics underscore the representativeness of our sample, offering a robust basis for analyzing the relationship between digital orientation and SME performance across different sectors and regions.

Table 5 Sample distribution: countries
Full size table
Table 6 Sample distribution: age (categorical)
Full size table
Table 7 Sample distribution: industries
Full size table

5 Results

5.1 Descriptive results

The correlations among all variables are displayed in Table 8. The correlation table shows relatively strong correlations among the individual subdimensions and the overall digital orientation variable, which is expected as all independent variables are based on one construct. Consequently, we estimate the dimensions in separated models each with all other explanatory variables. Overall, there is substantial independence among the control- and the independent variables which does not raise any multicollinearity concerns. Nevertheless, we tested the variance inflation factor (VIF) of all our independent variables in the separated models and found that none of them were close to the critical threshold of 2.5 (Johnston et al., 2018).

Table 8 Correlation matrix
Full size table
Table 9 Quadratic regression — digital orientation and organizational performance
Full size table
Table 10 Quadratic regression — digital orientation dimensions and organizational performance
Full size table

5.2 Estimation results

Tables 9 and 10 illustrate the results of the quadratic regression. We use six models to test our set of hypotheses. Model 1 illustrates the effect of digital orientation on SME growth without considering any control variables. We follow this procedure to initially estimate the unadjusted effect of digital orientation on SME performance. Afterward, we proceed with adjusted effects only, always considering all control variables. Model 2 contains the independent variable digital orientation in combination with the control variables. Model 3–6 show the independent variables: architecture configuration, capabilities, ecosystem coordination, and technology scope in combination with the control variables.

Fig. 1
figure 1

U-shape relationship between digital orientation and SME performance

Full size image
Fig. 2
figure 2

U-shape relationship between the individual dimensions of digital orientation and SME performance

Full size image

H1 predicts a U-shaped relationship between the overall construct of digital orientation of an SME and performance. This relationship is supported by Model 2, which shows significant estimates for both the linear term (\(\beta \) = \(-\)0.007; \(p<\) 0.05) and the squared term (\(\beta \) = 0.00005; \(p<\) 0.05). The relationship is illustrated in Fig. 1.

H2-H5 propose a U-shaped relationship between the individual subdimensions (architecture configuration, capabilities, ecosystem coordination, and technology scope) of digital orientation and SME performance. We find support for H2 with significant effects for both the linear term (\(\beta \) = \(-\)0.020; \(p<\) 0.01) and the squared term (\(\beta \) = 0.0002; \(p<\) 0.01) of the architecture configuration subdimension in Model 3. H3 suggests the existence of a U-shaped relationship between the subdimension capabilities and organizational performance. The results show support for this hypothesis, as both the linear term (\(\beta \) = \(-\)0.013; \(p<\) 0.05) and the squared term (\(\beta \) = 0.0002; \(p<\) 0.1) are significant. We find no support for the subdimension ecosystem coordination and therefore reject H4. H5 predicts a U-shaped relationship between technology scope and SME performance, which can be accepted as both the linear term (\(\beta \) = \(-\)0.016; \(p<\) 0.1) and the squared term (\(\beta \) = 0.0004; \(p<\) 0.05) are significant. All relationships of the subdimensions are illustrated in Fig. 2.

To validate the presence of our hypothesized U-shaped relationships, we performed the robustness test proposed by Haans et al. (2016) and Lind and Mehlum (2010), which involves testing the appropriate sign of \(\beta ^2\), controlling the cubic term model fit, and determining whether the inflection point of the U-shape lies within the data range using the 90% confidence interval. The U-shaped relationships for H1, H2, H3 and H5 are robust to all tests.

6 Discussion and conclusion

6.1 Theoretical implications

This study, grounded in the theoretical framework of RBV and DCs, empirically demonstrates a U-shaped curvilinear relationship between digital orientation and SME performance. While it is unclear how SMEs can drive digital transformation with limited resources (Barney, 1991; Li et al., 2018), our findings indicate that higher levels of SME performance are linked to both low and high ends of the digital orientation spectrum. Initially, pursuing a digital orientation tends to result in performance losses for SMEs, likely due to the resource and capability adjustments required. However, as digital orientation intensifies and becomes more ingrained, it positively impacts SME performance. This suggests that the strategic management of capabilities and resources linked to the setup of a digital orientation is crucial in the digital transformation journey of SMEs.

Consequently, our study aligns with the view of recognizing digital orientation as a valuable intangible capability (Kindermann et al., 2021; Schweiger et al., 2019; Teece, 2007) but also extends this notion by identifying the unique U-shaped relationship with SME performance. This contrasts with other studies who primarily focus on linear relationships (Wang et al., 2020), thus highlighting the complexity and dynamic nature of digital resources and capabilities in SMEs. Additionally, our findings offer a novel perspective in the ongoing discourse about the impact of digital orientation on organizational performance. Previous studies challenge the role of digital orientation in impacting performance (Calderon-Monge & Ribeiro-Soriano, 2023; Kraus et al., 2023; Nasiri et al., 2022), suggesting that despite the potential for digital orientation to mitigate operational difficulties and enable swift implementation of digital solutions (Arias-Pérez, 2022; Berman et al., 2023; Vial, 2019), it does not always lead to increased performance. A commonality between this and previous studies, as emphasized by Calderon-Monge and Ribeiro-Soriano (2023), is the understanding of digital maturity as a mediating factor between digital orientation and organizational performance. This aligns with our notion of the initial challenges faced by SMEs in digital orientation. Hence, we refine previous studies as we suggest that the mere adoption of digital technologies is not sufficient; rather, it is the depth and maturity of digital integration that ultimately drives performance (Calderon-Monge & Ribeiro-Soriano, 2023).

The divergence of insights from previous research may be due to the unique challenges and opportunities faced by SMEs in the digital transformation process, which can result in different patterns of performance compared to larger corporations (Eller et al., 2020; Hess et al., 2016; Mithas et al., 2012; OECD, 2021). Our study reveals a U-shaped pattern in SMEs, which might be attributed to the unique combination of agility and resource challenges faced by SMEs, a consequence of the liability of smallness (Aldrich & Auster, 1986) and their inherent DC (Li et al., 2018). This underscores the interplay between resource constraints and agility in shaping SME performance in the digital era, which is shaped by rapidly changing (digital) environments (Felicetti et al., 2024; Kraus et al., 2022). In this era however, it initially seems contradictory to have significant effects on both ends of the spectrum, as one would assume that the renunciation of any digital orientation would have severe consequences for SMEs. In this regard, previous studies have found that SMEs indeed are lagging behind in the process of digital transformation (OECD, 2021; Rupeika-Apoga et al., 2022). This observation is particularly intriguing from a RBV perspective (Barney, 1991). While studies suggest that SMEs are generally lagging in digital transformation, our findings imply that SMEs might be strategically choosing their level of digital orientation. This selective strategy could be resource-based, by balancing the costs and benefits of digital transformation in a resource-constrained environment.

SMEs often encounter significant resource limitations, which are frequently cited as major impediments to the implementation of new technology (Dörr et al., 2023). This aligns with our findings where the initial resource investment in digital orientation may strain SMEs overall resources, leading to a temporary dip in performance. The relevance of these constraints highlights the critical nature of strategic planning and resource management in the early stages of digital orientation for SMEs. Furthermore, our findings on the U-shaped relationship between digital orientation and SME performance resonate with insights on the sensing aspect of digital transformation (Leso et al., 2023). In SMEs, this sensing, which encompasses digital opportunity scanning and external exploration (Teece, 2007), may initially lead to resource allocation challenges, reflected in the initial dip the U-shaped curve. However, as SMEs progressively evaluate and adapt to digital opportunities, they likely begin to experience the performance benefits, aligning with the latter rise in the U-shaped curve. This nuanced understanding of SMEs’ digital strategies sets the stage for a deeper exploration of the underlying dynamics.

Thus, our findings enrich the discourse within the RBV (Barney, 1991) and DC (Teece, 2007) frameworks, particularly in the context of digital entrepre-neurship (Berman et al., 2023; Davidson & Vaast, 2010; Kraus et al., 2019) of SMEs. By demonstrating that digital orientation is not only a relevant but a critical driver of SME performance, our study counters the notion of a one-dimensional impact of digital transformation. It underscores the importance of considering the unique strategic approaches of SMEs in the digital era, thereby contributing to a more comprehensive understanding of how digital orientation influences organizational performance. Hence, SMEs need to critically think about their digital transformation, especially considering their resource investments to ultimately build their capabilities, and to not risk falling short of their competitive advantage (Schneider et al., 2023).

Furthermore, our findings contribute to the existing literature on strategic management by aligning with the perspective of researchers who recognize digital orientation as an emerging strategic orientation (Kane et al., 2015; Kindermann et al., 2021). By demonstrating that the pursuit of a digital orientation can be seen as a competitive advantage for organizations (Quinton et al., 2018), more specifically SMEs, we further fuel the scholarly discussion. Apart from the debate about whether digital orientation really is a new strategic orientation, there is another controversial discussion regarding the ability of SMEs to adopt and implement digital technologies (Dörr et al., 2023; Eller et al., 2020; Nguyen et al., 2015). Previous research showed that SMEs might perform worse in the successful adoption and integration of digital technologies due to their risk aversion (Ahmad et al., 2014; Taylor & Murphy, 2004). In this regard, our findings shed light on the concern of SMEs regarding the adoption and integration of digital technologies, confirming the short-term performance losses observed in prior research.

Apart from the empirical contribution to the explanation of the relationship between digital orientation and SME performance, we also contribute methodologically by analyzing the effect via text-based content analysis based on SMEs website-texts. This differs from many previous studies, which have relied on survey data or case studies to investigate digital transformation. The use of text-based content analysis allows for a more objective and scalable measurement of digital orientation, which could lead to new insights and extend the scope of the current literature on digital orientation and SME performance (Ungerer et al., 2021). Accordingly, the utilization of CATA opens up innovative methodological avenues for entrepreneurship research (He et al., 2023; Obschonka et al., 2020; von Bloh et al., 2020). Furthermore, the robustness and growing utilization of content analysis, as reviewed across organizational studies (Duriau et al., 2007), underscore its relevance and applicability in entrepreneurship research. The reliability and objectivity of content analysis (Elo et al., 2014), align with our aim to provide a more reliable picture of reality in understanding digital orientation’s impact on performance. This not only widens the ongoing discussion from a new perspective but also echoes the cross-disciplinary validation and innovative applications of CATA and content analysis, strengthening our methodological approach and its potential contribution to the field with a more reliable picture of reality (Hossnofsky & Junge, 2019).

6.2 Practical implications

The insights of this study have implications for both business management and policy. First, the study has shown that pursuing a digital orientation is essential when it comes to the successful development of SMEs in terms of performance in the digital era. Consequently, SMEs should reflect on their own resources and capabilities considering digital orientation and adapt to the prevailing trends, as this can ultimately lead to improved performance. SMEs that get “stuck in the middle” in the process of digital transformation, and thus in their digital orientation, do not show improved performance. On the contrary, these SMEs perform worst because the initial resource investments have already been made and the returns are still pending. In this situation, our findings underscore the importance of appropriate persistence with regard to the transformation process and the continuous intensification of digital orientation. Likewise, our results sensitize to the fact that the suitable degree of digital orientation highly depends on the respective business context. However, it is also important for SMEs to invest resources in digital orientation depending on their industry and their current state of digital transformation. Although investing resources in digital orientation has the potential to improve organizational performance, it should be noted that not every resource investment in this area is necessary, and the benefits may not be immediately reflected in performance. SMEs should therefore carefully assess the feasibility of pursuing digital orientation based on their specific context. Additionally, they should consider their willingness to undertake the necessary steps and resource investments required for successful implementation in the long-term.

Furthermore, the digital orientation of SMEs should be considered an essential prerequisite for building an ecosystem of competitive and sustainably viable SMEs within the digital era (Felicetti et al., 2024; Kraus et al., 2022). To improve the digital orientation of SMEs at the policy level, greater emphasis should be placed on promoting support programs that reflect the increasing importance of addressing the shortages of resources and capabilities of SMEs that have initiated and are persistently pursuing their digital transformation. To do so policymakers must pivot toward enabling a more robust digital infrastructure, by crafting and implementing support mechanisms that specifically address the resource and capability gaps hindering SMEs in their digital transformation endeavors. This encompasses financial incentives such as grants and subsidies to mitigate the initial costs of the adoption of digital technologies. Equally important is it to promote the digital skills among SME employees, ensuring that the human capital aligns with the technological advancements. This could be facilitated by providing access to special courses to develop digital skills. In line with the RBV, these courses should be as low-threshold, early and, above all, as cost-efficient as possible to enable broad application among SMEs.

6.3 Limitations and future research implications

This study is subject to limitations that open up opportunities for future investigations. One of the main limitations is the cross-sectional nature of our study, which can only provide a snapshot of the digital orientation of SMEs at a certain point in time. Although this approach offers valuable insights, it is unable to capture the dynamic and ongoing nature of digital transformation. Although we mitigated the limitations of cross-sectional data by using developmental variables, future research could benefit from a longitudinal design to fully explore the temporal dynamics and potentially curvilinear relationships in the digital transformation process in SMEs. Such a design would allow for a more nuanced understanding of how changes in digital orientation over time relate to changes in organizational performance. Additionally, this study relies on a CATA-based measurement of digital orientation through the analysis of website-texts, which draws on existing research of strategic posture (Grühn et al., 2017; Kindermann et al., 2021; Short et al., 2018). Therefore, the variable creation results from the communicated degree of digital orientation based on the Sapir-Whorf hypothesis and the attention SMEs devote to the topic (Abrahamson & Hambrick, 1997). However, organizations may not always see the need to directly and sometimes also may not subtly communicate strategic orientation through external communication channels such as websites. Therefore, future research could supplement secondary text data with primary data that directly captures the degree of digital orientation in an organization, to complement and complete the coverage of the degree of digital orientation.

Another limitation is the time frame in which we created our set of variables. Our set of variables is based on the years 2019–2022, which could be directly affected by the COVID-19 pandemic. Although we cannot control for the actual impact of the crisis, it applies equally to all our variables and SMEs in our dataset, mitigating this effect. For this reason, and because we wanted to examine a temporal relationship between the variables in our study as recent as possible, we opted for the most recently available performance data.

We also note the limitation of our current set of growth indicators and the potential for their reevaluation or expansion to provide a more robust link between digital orientation and organizational growth. Furthermore, we focus exclusively on the organizational level. Future research should examine more closely how digital orientation interacts with the skills and characteristics of decision makers, i.e., individuals, or within teams (Bamela et al., 2022). Since SME executives in particular tend to exert excessive influence on their organizations (Marcati et al., 2008), personal attitudes may have a major impact on the extent and rate of progress of digital orientation and digital transformation within SMEs and thus on organizational performance. Finally, although we control for the manufacturing industry, future research should take a more nuanced look at other industries and regional differences to examine the respective effects and consider industry and region-specific trends (Hossnofsky & Junge, 2019).

6.4 Conclusion

This study enriches the ongoing discourse on digital transformation and organizational performance by considering the relationship between digital orientation and the performance of SMEs from both a RBV and DC perspective. Our findings reveal a curvilinear U-shaped relationship between digital orientation and SME performance. SMEs at both ends of the spectrum demonstrate increased performance. The initial negative impact of pursuing a digital orientation diminishes as the intensity of the orientation increases. These results bear substantial implications for SMEs navigating the disruptive landscape of the digital era. Our study underscores that while the resource investments of pursuing a digital orientation may present initial barriers, the long-term performance benefits for SMEs outweigh these initial challenges. SMEs should therefore ensure that they are not “stuck in the middle” in terms of their digital orientation. Furthermore, our study emphasizes the necessity for SMEs to cultivate a strategic approach to digital transformation. Digital orientation only has a beneficial impact on organizational performance if the transformation process has either taken place at a very low level or is being consistently intensified toward a high degree of digital orientation. In conclusion, our study offers novel perspectives for digital entrepreneurship research, specifically on the relationship between digital orientation and organizational performance. It serves as a starting point for future research to further explore and develop this relationship in broader contexts and settings.