To cope with change and achieve sustainable, high-quality development, companies must implement digital transformation. Digital tools are critical for companies to remain competitive and successful. This study analyzes the factors that influence the digital transformation of Moroccan companies by contextualizing the technology–organization–environment (TOE) model. It seeks to examine the explanatory scope of this theoretical framework in the context of developing countries and to identify any potential limitations of its applicability.
A quantitative approach based on the partial least squares structural equation modeling method was used to test the conceptual model. The data were collected through a structured questionnaire addressed to professionals from Moroccan companies engaged in or considering digital transformation.
The results highlight the significant contribution of several technological and environmental factors. On the other hand, certain organizational dimensions, such as company size, digital cost or organizational readiness, proved to be insignificant, which calls into question the universality of the TOE model in emerging markets.
Unlike existing research that applies the TOE model in a standard manner, this study offers a critical and contextualized reading of its relevance in the Moroccan socio-economic context. It thus contributes to enriching theoretical debates and proposes avenues for adapting analytical frameworks or mobilizing hybrid methodologies in future research.
Introduction
The adoption of innovative technology throughout enterprises and organizations is known as Industry 4.0 (Nimmi, Vilone, & Jagathyraj, 2021). It focuses on automating processes and is defined by the smartness of IT systems that combine big data, artificial intelligence, etc. (Klingenberg, Borges, & do Vale Antunes, 2022). Digitization has started to spread in many nations as the key to competitiveness for companies of all sizes with the emergence of the later, digital transformation should be understood as an evolutionary and organization-wide process that extends beyond technology adoption to include strategic alignment, organizational restructuring and capability development (Omol, 2024).
Researchers have conducted several studies on companies' adoption of innovation. Nguyen, Le, and Vu (2022) demonstrated the motivations for companies in Vietnam to adopt online retailing (ORE). The findings reveal that elements such as observability are related to the technological context. Additionally, factors such as entrepreneurial orientation play a role in the environmental and organizational contexts, including government support, perceived trends and legal frameworks that aid ORE adoption. According to Putra and Santoso (2020), key factors in Indonesian businesses' adoption of e-business include organizational context (e.g. management support, financial resources and innovation), technology context (e.g. observability, complexity, compatibility and perceived benefits) and environmental factors (e.g. supplier support and government support).
Some of the models that have been used to explain how businesses adopt innovation include the technology–organization–environment (TOE) framework (Tornatzky, Tchell, & Alok, 1990), the Unified Theory of Technology Acceptance and Use (UTAUT) (Salimon et al., 2021) and the technology acceptance model (TAM). These models examine a range of internal and external business components using methods derived from system analysis (Chatterjee, Rana, Dwivedi, & Baabdullah, 2021). Academics frequently use the TOE framework, one of the most popular theoretical frameworks, in their research on business IT adoption.
For businesses trying to increase their competitiveness and fit the needs of the worldwide market, digital transformation is now a strategic lever. Although many studies have concentrated on the drivers of this change, few have particularly investigated the setting of Moroccan businesses, considering their structural, economic, and cultural particularities. Beyond examining adoption drivers, this study critically assesses the external validity of the TOE framework in an emerging economy context. TOE has, for the most part, received validation in developed economies, but in the most constrained, institutionally fragile, SME-dominated environments, its applicability remains largely uncharted. This research therefore contributes by identifying the contextual boundary conditions under which the TOE model operates in the Moroccan setting.
Tornatzky et al. (1990) claimed that this model offers a good understanding of the application of new technology. Its approach to digital adoption is broad and adaptable, considering the organizational environment and the technological settings. Many studies have confirmed this method (Chatterjee et al., 2021; Abed, 2020), they examine digital adoption. This article evaluated the most often used technique, the TOE framework, for deciding whether companies should adopt new technologies. TOE academics believe it to be legitimate, reliable and correct, given its obvious internal and external structures. The altered innovation model was examined in this work using partial least squares structural equation modeling (PLS-SEM).
Unlike most studies conducted in developed countries, which confirm the generalized validity of the TOE model, our results show that certain dimensions, particularly organizational and environmental ones (such as the cost of digital technology or company size), do not have a significant effect in the Moroccan context. This finding suggests that the applicability of the TOE model may be limited in certain specific institutional or structural environments. Without claiming to offer a contextualized version of the model, our study highlights the need to consider a future adaptation of the TOE framework to better reflect the realities of emerging economies, characterized by the predominance of SMEs and persistent institutional constraints.
The study mobilizes the PLS-SEM method, which is widely proven, but here contextualized to analyze the robustness of the relationships between technological, organizational and environmental factors and digital adoption.
The two research questions are as follows:
Which of the following are the most important technical, organizational and environmental reasons for companies to adopt digital transformation?
What practical and theoretical lessons can be drawn for public decision-makers and business leaders?
This study will examine factors that affect Moroccan firms' decisions to go digital. It will use a well-framed and well-defined method. Based on the literature, we divide this work into several sections. The next section reviews the theory. It includes a literature review on adopting digital transformation. The section also outlines our research framework and postulates derived from TOE theory. Section 3 details our research methodology. Data analysis and results are presented in Section 4, and a discussion is provided in Section 5. It concludes with conclusions and suggestions.
Literature review
Research model
Tornatzky et al. (1990) created the TOE framework, which stands for “technology, organization and environment,” to analyze the factors that allow businesses to use digital technologies (Fernando, 2021). While the organizational context discusses topics like firm size and available resources, the technical context focuses on the tools that organizations want both now and in the future. According to Tornatzky et al. (1990), the environmental context examines the larger environment in which businesses function, including governmental regulations. According to Jovic, Tijan, Vidmar, and Pucihar (2022), a variety of organizational, technological and environmental factors significantly impact how companies adapt to digital change.
Beyond identifying adoption factors, this research model is designed to critically assess the explanatory power of the TOE framework in an emerging economy context. By empirically testing technological, organizational and environmental variables within Moroccan firms, the model allows us to identify which dimensions retain their significance and which lose explanatory strength under specific institutional and structural constraints. This research provides a model that highlights the essential elements driving organizations, as indicated in Figure 1.
The conceptual model consists of rectangular boxes arranged into three contextual groups and a central outcome box. On the far left, a large rectangular box is labeled “Technology context” and contains three vertically stacked boxes with rounded corners: “Compatibility”, “Digital employee”, and “Relative advantage”. In the bottom center, a large rectangular box is labeled “Environment context” and contains four vertically stacked boxes with rounded corners: “Competitive pressure”, “Government support”, “Partner pressure”, and “Legal framework”. On the far right, a large rectangular box is labeled “Organizational context” and contains four vertically stacked boxes with rounded corners: “Digital cost”, “Firm size”, “Organizational readiness”, and “Top management support”. At the top center, a rectangular box with rounded corners is labeled “Digital transformation adoption”. Regarding the arrows: A horizontal rightward arrow labeled “H 1 a, H 2, H 3” points from the “Technology context” box to “Digital transformation adoption”. A horizontal leftward arrow labeled “H 4, H 5, H 6, H 7” points from the “Organizational context” box to “Digital transformation adoption”. A vertical upward arrow labeled “H 8 a, H 9, H 10, H 11” points from the “Environment context” box to “Digital transformation adoption”. Additionally, a horizontal rightward arrow labeled “H 1 b” points from “Relative advantage” to “Top management support”, and a horizontal rightward arrow labeled “H 8 b” points from “Competitive pressure” to “Top management support”.Conceptual model
The conceptual model consists of rectangular boxes arranged into three contextual groups and a central outcome box. On the far left, a large rectangular box is labeled “Technology context” and contains three vertically stacked boxes with rounded corners: “Compatibility”, “Digital employee”, and “Relative advantage”. In the bottom center, a large rectangular box is labeled “Environment context” and contains four vertically stacked boxes with rounded corners: “Competitive pressure”, “Government support”, “Partner pressure”, and “Legal framework”. On the far right, a large rectangular box is labeled “Organizational context” and contains four vertically stacked boxes with rounded corners: “Digital cost”, “Firm size”, “Organizational readiness”, and “Top management support”. At the top center, a rectangular box with rounded corners is labeled “Digital transformation adoption”. Regarding the arrows: A horizontal rightward arrow labeled “H 1 a, H 2, H 3” points from the “Technology context” box to “Digital transformation adoption”. A horizontal leftward arrow labeled “H 4, H 5, H 6, H 7” points from the “Organizational context” box to “Digital transformation adoption”. A vertical upward arrow labeled “H 8 a, H 9, H 10, H 11” points from the “Environment context” box to “Digital transformation adoption”. Additionally, a horizontal rightward arrow labeled “H 1 b” points from “Relative advantage” to “Top management support”, and a horizontal rightward arrow labeled “H 8 b” points from “Competitive pressure” to “Top management support”.Conceptual model
Three tiers make up the model: technological, organizational and environmental. The environmental context consists of a combination of the following measurable components: the digital employees, relative advantages and compatibility within the technological context; organizational readiness, digital costs, firm size and top management support within the organizational context; partner pressure, competitive pressure, government support and the legal framework within the environmental context. The study shows a number of factors positively influencing decision-making among Moroccan businesses on the digital transformation.
The structural model consists of twelve circles and several rectangular boxes that indicate statistical relationships. In the center, a large circle is labeled “D T A” and contains the value “0.657”. Three arrows point upward from “D T A” to three rectangular boxes arranged horizontally as “D T A 1”, “D T A 2”, and “D T A 3” with labels “0.867 (0.000)”, “0.874 (0.000)”, and “0.874 (0.000)”, respectively. To the left, three circles are arranged vertically. The top circle is labeled “C P T”, with two rectangular boxes “C P T 1” and “C P T 2” arranged vertically and both labeled “0.909 (0.000)”. The middle circle is labeled “D E”, with four rectangular boxes “D E 1” through “D E 4” arranged vertically and labeled “0.742 (0.000)”, “0.706 (0.000)”, “0.727 (0.000)”, and “0.755 (0.000)”, respectively. The bottom circle is labeled “R A”, with four rectangular boxes “R A 1” through “R A 4” arranged vertically and labeled “0.859 (0.000)”, “0.886 (0.000)”, “0.882 (0.000)”, and “0.877 (0.000)”, respectively. To the right, three circles are arranged vertically. The top circle is labeled “D C”, with three rectangular boxes “D C 1” through “D C 3” arranged vertically and labeled “0.832 (0.000)”, “0.860 (0.000)”, and “0.772 (0.000)”. The middle circle is labeled “F I S”, with three rectangular boxes “F I S 1” through “F I S 3” arranged vertically and labeled “0.853 (0.000)”, “0.842 (0.000)”, and “0.810 (0.000)”. The bottom circle is labeled “O R R”, with three rectangular boxes “O R R 1” through “O R R 3” arranged vertically and labeled “0.861 (0.000)”, “0.875 (0.000)”, and “0.873 (0.000)”. At the bottom, four circles are arranged horizontally. From left to right: “L F” has two arrows pointing downward to two rectangular boxes arranged horizontally and labeled “L F 1” and “L F 2” with arrow values “0.877 (0.000)” and “0.893 (0.000)”. The circle “P P” has two arrows pointing downward to two rectangular boxes arranged horizontally, labeled “P P 1” and “P P 2” with arrow values “0.962 (0.000)” and “0.964 (0.000)”. The circle “G S” has two arrows pointing downward to two rectangular boxes arranged horizontally, labeled “G S 1” and “G S 2” with arrow values “0.917 (0.000)” and “0.918 (0.000)”. The circle “C P” has three arrows pointing downward to three rectangular boxes arranged horizontally, labeled “C P 1”, “C P 2”, and “C P 3” with arrow values “0.859 (0.000)”, “0.830 (0.000)”, and “0.861 (0.000)”. To the far right, a circle labeled “T M S” contains the value “0.478” and has four arrows pointing rightward to four rectangular boxes arranged vertically labeled “T M S 1” through “T M S 4” with arrow values “0.783 (0.000)”, “0.821 (0.000)”, “0.803 (0.000)”, and “0.784 (0.000)”. Regarding the relationships between circles: A solid arrow labeled “0.005” points from “C P T” to “D T A”. A solid arrow labeled “0.592” points from “D E” to “D T A”. A solid arrow labeled “0.003” points from “R A” to “D T A”. A solid arrow labeled “0.010” points from “I F” to “D T A”. A solid arrow labeled “0.000” points from “P P” to “D T A”. A solid arrow labeled “0.016” points from “G S” to “D T A”. A solid arrow labeled “0.004” points from “C P” to “D T A”. A solid arrow labeled “0.020” points from “T M S” to “D T A”. A solid arrow labeled “0.383” points from “O R R” to “T M S”. A solid arrow labeled “0.901” points from “F I S” to “D T A”. A solid arrow labeled “0.902” points from “D C” to “D T A”. Additionally, a solid arrow labeled “0.010” points from “R A” to “T M S”. A solid arrow labeled “0.000” points from “C P” to “T M S”.Path coefficient for structural model
The structural model consists of twelve circles and several rectangular boxes that indicate statistical relationships. In the center, a large circle is labeled “D T A” and contains the value “0.657”. Three arrows point upward from “D T A” to three rectangular boxes arranged horizontally as “D T A 1”, “D T A 2”, and “D T A 3” with labels “0.867 (0.000)”, “0.874 (0.000)”, and “0.874 (0.000)”, respectively. To the left, three circles are arranged vertically. The top circle is labeled “C P T”, with two rectangular boxes “C P T 1” and “C P T 2” arranged vertically and both labeled “0.909 (0.000)”. The middle circle is labeled “D E”, with four rectangular boxes “D E 1” through “D E 4” arranged vertically and labeled “0.742 (0.000)”, “0.706 (0.000)”, “0.727 (0.000)”, and “0.755 (0.000)”, respectively. The bottom circle is labeled “R A”, with four rectangular boxes “R A 1” through “R A 4” arranged vertically and labeled “0.859 (0.000)”, “0.886 (0.000)”, “0.882 (0.000)”, and “0.877 (0.000)”, respectively. To the right, three circles are arranged vertically. The top circle is labeled “D C”, with three rectangular boxes “D C 1” through “D C 3” arranged vertically and labeled “0.832 (0.000)”, “0.860 (0.000)”, and “0.772 (0.000)”. The middle circle is labeled “F I S”, with three rectangular boxes “F I S 1” through “F I S 3” arranged vertically and labeled “0.853 (0.000)”, “0.842 (0.000)”, and “0.810 (0.000)”. The bottom circle is labeled “O R R”, with three rectangular boxes “O R R 1” through “O R R 3” arranged vertically and labeled “0.861 (0.000)”, “0.875 (0.000)”, and “0.873 (0.000)”. At the bottom, four circles are arranged horizontally. From left to right: “L F” has two arrows pointing downward to two rectangular boxes arranged horizontally and labeled “L F 1” and “L F 2” with arrow values “0.877 (0.000)” and “0.893 (0.000)”. The circle “P P” has two arrows pointing downward to two rectangular boxes arranged horizontally, labeled “P P 1” and “P P 2” with arrow values “0.962 (0.000)” and “0.964 (0.000)”. The circle “G S” has two arrows pointing downward to two rectangular boxes arranged horizontally, labeled “G S 1” and “G S 2” with arrow values “0.917 (0.000)” and “0.918 (0.000)”. The circle “C P” has three arrows pointing downward to three rectangular boxes arranged horizontally, labeled “C P 1”, “C P 2”, and “C P 3” with arrow values “0.859 (0.000)”, “0.830 (0.000)”, and “0.861 (0.000)”. To the far right, a circle labeled “T M S” contains the value “0.478” and has four arrows pointing rightward to four rectangular boxes arranged vertically labeled “T M S 1” through “T M S 4” with arrow values “0.783 (0.000)”, “0.821 (0.000)”, “0.803 (0.000)”, and “0.784 (0.000)”. Regarding the relationships between circles: A solid arrow labeled “0.005” points from “C P T” to “D T A”. A solid arrow labeled “0.592” points from “D E” to “D T A”. A solid arrow labeled “0.003” points from “R A” to “D T A”. A solid arrow labeled “0.010” points from “I F” to “D T A”. A solid arrow labeled “0.000” points from “P P” to “D T A”. A solid arrow labeled “0.016” points from “G S” to “D T A”. A solid arrow labeled “0.004” points from “C P” to “D T A”. A solid arrow labeled “0.020” points from “T M S” to “D T A”. A solid arrow labeled “0.383” points from “O R R” to “T M S”. A solid arrow labeled “0.901” points from “F I S” to “D T A”. A solid arrow labeled “0.902” points from “D C” to “D T A”. Additionally, a solid arrow labeled “0.010” points from “R A” to “T M S”. A solid arrow labeled “0.000” points from “C P” to “T M S”.Path coefficient for structural model
However, its application in specific contexts, like that of Morocco, requires adaptation. We have integrated technological variables (digital infrastructure, data security), organizational variables (company size, organizational readiness, digital leadership) and environmental variables (competitive pressure, government support, availability of skills).
Cherkaoui (2024) state that Moroccan companies' digitalization is obstructed by organizational obstacles aligned with managerial conduct, a lack of institutional pressure and a predominance of SMEs. However, empirical studies that have analyzed these phenomena using the TOE framework are very few. Indeed, a major theoretical question still exists: in a Moroccan context, factors that Western theorists have considered to be important (e.g. the size of the company or the price of digital technologies) may not explain the situation similarly. This is because of the strong lack of financing, a still embryonic regulatory framework and overreliance on international financing.
El Amrani and al (2023) pointed out the fragmented nature of Moroccan digital transformation literature, documenting, inter alia, the lack of digital infrastructure and disparate regional distribution.
Taken together, these observations suggest that while the TOE framework is widely validated, its universal applicability remains questionable when applied to emerging economies with distinct structural and institutional characteristics. Our study aims to fill this dual gap by articulating these empirical findings around a robust theoretical framework – the TOE model – while demonstrating its explanatory limitations in the Moroccan context. The theoretical gap lies in the need to test the applicability of a model established in a distinct institutional and cultural environment, while the empirical gap concerns the lack of quantitative validation of these dynamics within Moroccan companies, with the majority of existing work remaining descriptive or qualitative.
Research hypothesis
Technological context
Relative advantages:
It is used to describe the degree of consistency that might result in improved business operations; it is the set of benefits a company can gain from digital technologies. It is the idea that an invention is regarded as more useful to a business than its predecessor (Rogers, 2003). Several studies have shown the competitive advantages of companies adopting digital transformation, such as improving business efficiency, helping to reduce and minimize costs (Yin, 2022) and transforming organizational business models. Based on these results, this proposal was suggested:
Relative advantage has a favorable influence on an organization's digital transformation.
Top management's support for business digitization is positively impacted by relative advantage.
Compatibility:
The degree to which an invention is deemed consistent with adoption criteria and contemporary values is known as compatibility (Mahakittikun, Suntrayuth, & Bhatiasevi, 2020; Alsetoohy, Ayoun, Arous, Megahed, & Nabil, 2019) demonstrated in his study that companies adopt innovations that only somewhat align with their existing values and only require little adjustments. (Khan, Khan, Bahadur, & Ali, 2021) focused on mobile payment systems as a digital technology; if they are compatible and adapted to the existing technological architecture, they can be adopted and used by companies, the following studies are suggested:
The adoption of digitalization by corporations is positively correlated with compatibility.
Digital employee:
Many businesses are starting to see the value of assembling a group of people who are digitally literate and possess the digital talent required to embrace digital technologies as part of the execution of a digital transformation strategy (Teng, Wu, & Yang, 2022). Enhancing digital abilities is regarded as a crucial production component and is necessary for initiating digital transformation (Li & Cao, 2022). Gabriel et al. (2020) assert that there is a shortage of digital talent and a lack of human capital, which automatically limits digital transformation in companies; thus, firms need to establish virtual human resources and employ resources for learning to be ready for the future (Bennett et al., 2021), so the organization's biggest challenge is not the business but rather the people. Consequently, the following studies are suggested:
The transition to new technologies is positively influenced by the presence of digitally skilled personnel.
Organizational context
Digital cost:
Refer to the charges companies incur when implementing measures related to digital transformation or new digital technologies. According to some researcher, the cost factor is important in adopting digital technologies and is measured in terms of administrative costs and financial investment. (Ainin, Parveen, Moghavvemi, Jaafar, & Mohd Shuib, 2015) emphasize the importance of profitability as a critical aspect of implementing digital technologies. Hence, this research presents the following hypotheses:
Digital costs negatively influence companies' decisions to adopt digitalization
Firm size:
Firm size is important for implementing a digital strategy since the larger the business, the more capable it is of embracing digital technologies (Riemenschneider, Harrison, & Mykytyn, 2003). Generally, large companies have more resources and capacity to invest in implementing these tools; however, small companies have fewer resources to ensure the burdens generated by the implementation of digitalization. Thus, this study puts up the following theories:
Company size influences companies' decisions to adopt digital transformation
Top management support:
Top management plays a crucial role in decision-making when implementing a digital strategy; this highlights the importance of company leaders in endorsing or opposing digital transformation. The extent to which executives adapt to the technological capabilities of innovations (Maroufkhani, Iranmanesh, & Ghobakhloo, 2023) defines the mindset of management to establish a viable atmosphere, provide support and ensure the availability of necessary resources. (Gui, Shaharudin, Mokhtar, Karmawan, & Suryanto, 2020; Abbasi, Rahim, Wu, Iranmanesh, & Keong, 2022). Therefore, we put up the following theory:
Organizations' adoption of digital transformation is impacted by top management support.
Organizational readiness:
Organizational readiness is the state of readiness of the organization and its readiness to implement digital technology, as well as the availability of the resources needed to succeed, including infrastructure, human resources and necessary financing; this is technical readiness for innovation (Chatterjee et al., 2021; Gui et al., 2020). Therefore, we put up the following theory:
Implementing digital transformation in firms is greatly facilitated by organizational readiness.
Environmental context:
Competitive pressure:
Competitive pressure is the impact on enterprises within the same line of industry (Wong, Leong, Hew, Tan, & Ooi, 2020). It is evidenced that competitive pressure is important in transforming companies, apart from being a determinant of the innovativeness and the adoption of new technologies. Competitive pressure has been shown to cause some organizations to make a shift toward digital transformational and, thereby, directly or indirectly cause the presence of top management support. Hence, this research posits the following hypotheses:
Competitive pressure enhances digital transformational adoption by firms.
Competitive pressure has an impact on top management support for digital transformational adoption by firms.
Government support
Government support is a set of initiatives to boost innovation and encourage businesses to adopt new technologies (Gui et al., 2020; El-Haddadeh, 2021). This support included training, funding and technical assistance. Financial and nonfinancial support from the government is critical for implementing new digital technologies. It directly encourages high-level executives to understand the benefits of big data analytics and encourages them to use it. If the government supports digital transformation and the implementation of new digital technologies, executives will react positively, which will increase the possibility of transformation. Consequently, we suggest this hypothesis:
Political support had a beneficial impact on the organization's digital transformation.
Partner pressure
Every organization has a relationship with its business partners, and it is necessary to consider this when implementing new technology. Many studies have shown that partners' demand and readiness are key to adopting innovations. The interactions with partners (Lin et al., 2008) affect innovation adoption. When partners have adopted innovation, it will positively impact the company's digital transformation. Pressure from business partners is crucial to implementing innovations. It creates a platform for sharing information among stakeholders, including suppliers and customers. Business leaders can explain this by increasing their interactions with partners (Chen, Yin, Browne, & Li, 2019). Thus, the following assumption:
Pressure from business partners affects companies' acceptance of digital transformation.
Legal framework
The legal framework comprises the rules and guidelines that corporations must follow to (Hiran & Henten, 2020). Companies may face limitations in adopting new digital tech to increase performance (Maroufkhani et al., 2023). According to this study, businesses that have clout and need government regulation are more inclined to use digital tools like big data. As a result, the adoption of digital transformation is greatly influenced by the regulatory environment. The following hypothesis results from this:
Businesses are encouraged to use digital transformation by the regulatory framework.
Materials and methods
For the data collection strategy, we used a qualitative method illustrated by a questionnaire survey. First, we developed a preliminary questionnaire, and then, based on feedback and recommendations from researchers and experts in the same field, we adjusted the questionnaire to create our final version.
The different characteristics were measured using a 5-point Likert scale (1 = severely disagree, 2 = disagree, 3 = neutral, 4 = agree, and 5 = extremely agree) to simplify the completion process for participants. There are 42 questions total on the survey, separated into two sections: the first one deals with our people's social and demographic information, and the second part is for measuring the variables within the framework of the TOE method; in the technical context, we find the factors relative advantage, compatibility and digital employee; in the organizational context, it consists of organizational readiness, firm size, digital cost and top management support; and the environmental context includes competitive pressure, government support, legal framework and partner pressure.
We contacted our target group: directors, managers, department heads and IT managers in various Moroccan companies. The team distributed the questionnaire online via Google Forms. They sent the survey link by email and on LinkedIn. We first tried to explain the study's context to our population. We defined digital transformation and said there were no right or wrong answers. But over three months, we collected 478 responses. After deleting incomplete answers, we used 356 for analysis. According to Chin (1998), researchers must meet the 10-fold rule for the least sample size in an SEM PLS study. It means to choose the greater of these two. The most independent variables should have 10 times as many measurement items, or the most endogenous variables should have ten times as many exogenous variables. In our case, the independent variables with the most elements are top management support (TMS) and digital employees (DE), with four each. Digital transformation adoption (DTA) has 11 exogenous variables. This proves that the 356 responses in this study meet the least sample size of over 110.
The use of the PLS-SEM method is justified by the exploratory nature of this research and the complexity of the model, which incorporates several latent variables and structural relationships. This approach is particularly suited to medium-sized samples (n = 356) and emerging contexts, where data do not always meet the assumptions of normality. It also provides robust estimates in models that include multiple hierarchical relationships between latent variables.
The questionnaire items were adapted from existing literature on the TOE model (Tornatzky & Fleischer, 1990) and contextualized for the Moroccan case. Each construct was measured using several items on a 5-point Likert scale (see Appendix 1). For example, the Relative Advantage dimension was assessed using statements such as: “RA1: The use of digital transformation makes companies more efficient. RA2: The use of digital transformation improves customer service. RA3: The use of digital transformation reduces costs.RA4: The adoption of digital transformation makes companies more competitive.”
Results
Demographic data
Table 1 presents the basic demographic information and characteristics of the respondents. First, the respondents included 228 men and 128 women; males represented the majority. More than 83% of the respondents held a master's degree (47.2%) or an undergraduate degree (36%). Most respondents fell within the age range of 31–35 (33.7%) and 36–40 (27%). Among the respondents, 37.4% had three to five years of expertise, while 36% had 10 or more years. Among the survey respondents, 32.6% were department managers, 24.2% were IT managers, 21.6% were general managers and 21.6% were others. Of the participating companies, over 53% are small- and medium-sized, with 10 to 300 employees. 43.5% are large, with over 300 employees. The remaining 3.1% are the smallest companies. We decided to consider the sizes and sectors of the companies. 31.5% were in manufacturing, 21.9% in automotive, 10.4% in trade, 13.2% in transport and storage, 11.2% in textiles and 11.8% in other sectors. The diversity of areas in our results is an advantage. It allows businesses to use this study to transform themselves, regardless of their sector.
Respondents' quantitative data
| Element | Category | Frequency | % |
|---|---|---|---|
| Gender | Male | 228 | 64.0 |
| Female | 128 | 36.0 | |
| Education | College and below | 54 | 15.2 |
| Undergraduate | 128 | 36.0 | |
| Master | 168 | 47.2 | |
| PhD | 6 | 1.7 | |
| Age | <25 years | 31 | 8.7 |
| 25–30 years | 77 | 21.6 | |
| 31–35 years | 120 | 33.7 | |
| 36–40 years | 96 | 27.0 | |
| >40 years | 32 | 9.0 | |
| Years of experience | <3 years | 36 | 10.1 |
| 3–5 years | 133 | 37.4 | |
| 6–10 years | 128 | 36.0 | |
| 11–15 years | 41 | 11.5 | |
| >15 years | 18 | 5.1 | |
| Sector of activity | Manufacturing industry | 112 | 31.5 |
| Textile industry | 40 | 11.2 | |
| Automotive industry | 78 | 21.9 | |
| Trade | 37 | 10.4 | |
| Transport and storage | 47 | 13.2 | |
| Other | 42 | 11.8 | |
| Job title | Managing director | 77 | 21.6 |
| IT Director | 86 | 24.2 | |
| Director/Department | 116 | 32.6 | |
| Other | 77 | 21.6 | |
| Number of employees | <10 | 11 | 3.1 |
| 10–199 | 61 | 17.1 | |
| 200–300 | 129 | 36.2 | |
| >300 | 155 | 43.5 |
Researchers tested for multicollinearity to look for strong correlations among the independent variables. High correlations can affect the model's estimation. So, we must test for multicollinearity before using SEM. According to the greatest variance inflation factor (VIF), it was 1.690, less than the five-point cutoff suggested (Hair, Risher, Sarstedt, & Ringle, 2019). In contrast, the tolerance values of the different constructs were within the allowed range (0.1 to 1). This showed that multicollinearity was not an issue in this study. Table 2 presents the results of the multicollinearity assessment.
Results of the multicollinearity test
| Tolerance | VIF | |
|---|---|---|
| Relative advantage | 0.724 | 1,380 |
| Compatibility | 0.685 | 1,461 |
| Digital employee | 0.592 | 1,690 |
| Digital cost | 0.668 | 1,497 |
| Firm size | 0.718 | 1,394 |
| Top management support | 0.699 | 1,431 |
| Organizational readiness | 0.689 | 1,452 |
| Competitive pressure | 0.768 | 1,302 |
| Government Support | 0.721 | 1,387 |
| Partner pressure | 0.759 | 1,317 |
| Legal framework | 0.811 | 1,233 |
Reliability and convergence validity
| Constructs | Items | Loadings | Cronbach's α | CR | AVE |
|---|---|---|---|---|---|
| Competitive pressure (CP) | CP1 | 0.859 | 0.808 | 0.887 | 0.723 |
| CP2 | 0.830 | ||||
| CP3 | 0.861 | ||||
| Compatibility (CPT) | CPT1 | 0.909 | 0.790 | 0.905 | 0.826 |
| CPT2 | 0.909 | ||||
| Digital cost (DC) | DC1 | 0.832 | 0.761 | 0.862 | 0.676 |
| DC2 | 0.860 | ||||
| DC3 | 0.772 | ||||
| Digital employee (DE) | DE1 | 0.742 | 0.713 | 0.823 | 0.537 |
| DE2 | 0.706 | ||||
| DE3 | 0.727 | ||||
| DE4 | 0.755 | ||||
| Digital transformation adoption (DTA) | DTA1 | 0.867 | 0.842 | 0.905 | 0.760 |
| DTA2 | 0.874 | ||||
| DTA3 | 0.874 | ||||
| Firm size (FIS) | FIS1 | 0.853 | 0.783 | 0.874 | 0.697 |
| FIS2 | 0.842 | ||||
| FIS3 | 0.810 | ||||
| Government Support (GS) | GS1 | 0.917 | 0.812 | 0.914 | 0.841 |
| GS2 | 0.918 | ||||
| Legal framework (LF) | LF1 | 0.877 | 0.724 | 0.878 | 0.783 |
| LF2 | 0.893 | ||||
| Organizational readiness (ORR) | ORR1 | 0.861 | 0.839 | 0.903 | 0.756 |
| ORR2 | 0.875 | ||||
| ORR3 | 0.873 | ||||
| Partner pressure (PP) | PP1 | 0.962 | 0.922 | 0.962 | 0.927 |
| PP2 | 0.964 | ||||
| Relative advantage (RA) | RA1 | 0.859 | 0.899 | 0.930 | 0.767 |
| RA2 | 0.886 | ||||
| RA3 | 0.882 | ||||
| RA4 | 0.877 | ||||
| Top management support (TMS) | TMS1 | 0.783 | 0.810 | 0.875 | 0.637 |
| TMS2 | 0.821 | ||||
| TMS3 | 0.803 | ||||
| TMS4 | 0.784 |
Discriminant validity
| Constructs | CP | CPT | DC | DE | DTA | FIS | GS | LF | ORR | PP | RA | TMS |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CP | 0.850 | |||||||||||
| CPT | 0.565 | 0.909 | ||||||||||
| DC | 0.648 | 0.596 | 0.822 | |||||||||
| DE | 0.643 | 0.628 | 0.663 | 0.733 | ||||||||
| DTA | 0.669 | 0.645 | 0.619 | 0.635 | 0.872 | |||||||
| FIS | 0.619 | 0.554 | 0.557 | 0.628 | 0.581 | 0.835 | ||||||
| GS | 0.609 | 0.579 | 0.576 | 0.585 | 0.628 | 0.593 | 0.917 | |||||
| LF | 0.592 | 0.570 | 0.585 | 0.547 | 0.623 | 0.557 | 0.536 | 0.885 | ||||
| ORR | 0.664 | 0.570 | 0.681 | 0.622 | 0.626 | 0.595 | 0.574 | 0.608 | 0.870 | |||
| PP | 0.602 | 0.535 | 0.566 | 0.604 | 0.654 | 0.541 | 0.581 | 0.544 | 0.582 | 0.963 | ||
| RA | 0.547 | 0.581 | 0.582 | 0.613 | 0.628 | 0.551 | 0.492 | 0.559 | 0.534 | 0.556 | 0.876 | |
| TMS | 0.612 | 0.641 | 0.679 | 0.643 | 0.656 | 0.521 | 0.570 | 0.536 | 0.620 | 0.575 | 0.604 | 0.798 |
Measurement model
Researchers examine the measurement model in PLS-SEM to assess the model quality, which comprises reliability (Cronbach's alpha), convergent validity, and discriminant validity. Every idea fulfilled the standards set by Hair et al. (2019) with Cronbach's alpha and composite reliability (CR) values exceeding 0.70. The fact that all ideas have factor loadings in the model above the recommended threshold of 0.70 and an average variance extracted (AVE) above 0.50 suggests convergent validity, the results are reported in Table 3.
Second, the discriminant validity given by the criteria of Fornell–Larcker and cross-loading can be assessed by comparing the correlation coefficients of the ideas with the square root of the AVE, aiming to guarantee the whole difference between the several latent variables was identified. Since the square root of each idea's AVE is larger than the inter-construct correlation coefficients, the statistics show that all ideas exhibit discriminant validity, fulfilling the criteria set forth by Fornell and Larcker (1981). The results are presented in Table 4.
Structural model
PLS-SEM structural models need to find three indicators – cross-validated redundancy (Q2), effect size (f2) and coefficient of determination (R2)—in order to forecast exogenous components (Hair et al., 2019). Derived from the coefficient of determination (R2), the coefficient of determination R2 runs from 0 to 1 and evaluates the accuracy of the prediction of the model (Hair et al., 2019). A higher R2 value indicates a better prediction of endogenous components. Given our circumstances, the DTA's R2 is 0.657 and its updated R2 is 0.646. TMS therefore meets the criteria with an R2 of 0.478 and an adjusted R2 of 0.475.
According to Hair et al. (2019), the Cohen effect size (f2) was used to assess how endogenous constructs were affected by the removal of the predictor construct. Small, medium and high impact's related f2 values are 0.02, 0.15 and 0.35, respectively. The f2 value for all our constructs is 0.15; thus, the f2 values for the CP and RA variables on MSD are 0.217 and 0.198, respectively, which account for the strong and moderate influences on MSD.
Hair et al. (2019) state that a structural model's predictive accuracy is acceptable if Q2 is greater than 0. Our model's Q2 results show a value of 0.634 for DTA and 0.478 for TMS, which is larger than zero, showing that the model has a favorable predictive correlation. It is necessary to analyze the structural model to check the suitability of the hypotheses, that is, whether they are accepted or rejected. Smart PLS software generally uses the bootstrap method with a sample size of 5,000. Below are the path coefficients (beta), associated p-values, and t-statistics. The structural relationships between the constructs and their corresponding path coefficients are illustrated in Figure 2.
The hypothesis testing results from Table 5 reveal that, among the 13 factors examined, digital employees, digital cost, firm size and organizational readiness are insignificant at 0.05. Competitive pressure, compatibility, government support, legal framework, partner pressure, relative advantage and top management support are significant. Relative advantage and competitive pressure influence top management support.
Hypothesis testing results
| Hypotheses | Relationship | Path coefficients | T statistics | p values | Result |
|---|---|---|---|---|---|
| H8a | CP → DTA | 0.154 | 2.892 | 0.004 | Supported |
| H8b | CP → TMS | 0.402 | 7.792 | 0.000 | Supported |
| H2 | CPT → DTA | 0.133 | 2.784 | 0.005 | Supported |
| H4 | DC → DTA | −0.007 | 0.124 | 0.902 | Not Supported |
| H3 | DE → DTA | 0.028 | 0.536 | 0.592 | Not Supported |
| H5 | FIS → DTA | 0.007 | 0.125 | 0.901 | Not Supported |
| H9 | GS → DTA | 0.117 | 2.414 | 0.016 | Supported |
| H11 | LF → DTA | 0.118 | 2.579 | 0.010 | Supported |
| H7 | ORR → DTA | 0.046 | 0.873 | 0.383 | Not Supported |
| H10 | PP → DTA | 0.170 | 3.658 | 0.000 | Supported |
| H1a | RA → DTA | 0.132 | 2.975 | 0.003 | Supported |
| H1b | RA → TMS | 0.384 | 8.406 | 0.000 | Supported |
| H6 | TMS → DTA | 0.124 | 2.318 | 0.020 | Supported |
Regarding the mediating effect, Table 6 shows a considerable correlation between the adoption of digital transformation, pressure from competitors and top management support, and another significant one between relative advantage and top management support and adoption of digital transformation.
Discussion
The TOE approach was employed in this study to identify the factors that have the most influence on businesses' decisions to digitalize or adopt a digital strategy. According to our analysis, the tech environment is driven by compatibility and relative advantage. Support from upper management and all environmental factors are necessary in the organizational setting. However, the decision is not influenced by factors like firm size, organizational preparedness, digital personnel or digital expenses.
The PLS-SEM findings reveal that RA (H1a) has a beneficial influence on organizations' embrace of digital transformation, as indicated by the probability results (p = 0.003 ≤ 0.05). This finding matches prior research. It showed that a major influence on firms' choice to adopt digital tools is their benefits. These tools include big data, AI (Chen, Li, & Chen, 2021), 3D printing, social media marketing (Abbasi et al., 2022), robotics and blockchain (Hashimy, Jain, & Grifell-Tatjé, 2022). These tools help companies reduce costs and become more efficient. Moreover, a relative advantage not only has a significant impact on the adoption of digital transformation but also positively influences top-level support for leadership (H1b), implying that the relative advantage generated by the use of digital technologies can strengthen top management support for going digital, according to research by Wong et al. (2020), top management support for digital transformation in Malaysian SMEs is bolstered by the advantages of digital technology.
The results (p = 0.005 ≤ 0.05) confirm that CPT (H2) has a significant impact on companies' decisions to adopt digital transformation; this agrees with (Maroufkhani et al., 2023; Tajudeen, Ainin, & Jaafar, 2018). Consistency between adoption needs and existing values is essential. Unlike previous studies by Teng et al. (2022) and Li and Cao (2022), which confirm a positive effect of employees' digital skills on adopting digitalization, this research found (p = 0.592 ≥ 0.05). Employees' digital capabilities (H3) have a limited effect on organizations' digital transformation; this happens because employees can gain digital skills over time, and organizations must ensure they keep improving their digital capabilities. Additionally, digital transformation may not integrate into production and could remain limited to other levels of the company. Li et al. (2022) explain this contradiction. Many companies have high turnover rates, including among digital workers; this can cause a lack of perseverance and limit their roles in digital transformation.
In the organizational context, digital cost (H4) does not have a significant impact on the adoption of digital transformation (p = 0.902 ≥ 0.05). This result contrasts with the findings of Ainin et al. (2015), who identified digital cost as a major determinant in implementing digital technologies. In the Moroccan context, this insignificance may reflect that many firms perceive digital investment not as a primary barrier but as a necessary expense for long-term competitiveness. Alternatively, it may indicate that cost considerations are outweighed by other organizational or strategic priorities in the decision to go digital. Second, company size (H5) does not have a significant impact on digital transformation adoption, as shown by the results obtained (p = 0.901 ≥ 0.05), which is in line with previous studies (Calof, 2020) and contradictory to studies by (Riemenschneider et al., 2003), who showed in their work that company size has a crucial impact on the use of digital technologies: the larger the organization, the more resources it has to adopt digital transformation and the smaller the company, the fewer resources it has to cover the costs of implementing digital tools. Companies can explain this contradiction by choosing to use less expensive digital tools that suit their needs and financial capabilities, such as social media.
The lack of significance of certain traditional factors in the TOE model, such as company size and digital costs, deserves special attention. Unlike the results observed in developed countries, these variables are not major determinants of digital adoption in Morocco. This specificity can be explained by the predominance of SMEs, the availability of public subsidies, and the managerial flexibility specific to the local context, where decisions are based more on the vision of the leader and institutional support than on organizational structure. These findings highlight the need to adapt the TOE model to emerging economies, taking into account their specific institutional and cultural mechanisms.
Also, top management support (H6) is a key component in the introduction of digitalization, reinforcing the research of (Abed, 2020; Vial, 2019), which demonstrated how important leadership is to the process of digital transformation and showed that it is an essential agent for accelerating the transformation process. It was discovered that top management's support for the company's digital transformation was positively influenced by the relative advantage (H1b) and competitive pressure (H8b) factors, indicating that senior managers must prioritize relative advantage and competitive pressure when adopting new technologies. Regarding the organizational readiness factor, we issued a hypothesis (H7) confirming that it has a favorable influence on the acceptance of the digital revolution; this is counter to the findings obtained in (Singh, Sharma, & Dhir, 2021) research, which confirms this contradiction and emphasizes that there is no relationship between organizational readiness and the choice to adopt digital transformation. This may be clarified (Wang, Liu, Liu, & Huang, 2022) by considering that it can be difficult for members of an organization who already possess rigid ways of thinking and acting to react to the novel innovations that digitization will generate. However, several studies have confirmed the opposite and that organizational readiness contributes significantly, as in the studies by (Wang et al., 2022; Gui et al., 2020).
One particularly interesting finding concerns the mediating role of top management support. The relationship between relative advantage and digital adoption is indirectly influenced by management support (H1b, H6 significant). This confirms that in Moroccan companies, where organizational culture remains highly centralized, digital initiatives only translate into actual adoption through the involvement and active support of management. This finding is consistent with research on transformational leadership and confirms that managers are the real catalysts for digitalization in environments where formal organizational structures are weak.
The third framework is the environmental one. All its factors boost the adoption of digital transformation; this starts with competitive pressure (H8a/H8b). The findings supported earlier research. They showed that competition boosts top management support and the digital revolution (Wong et al., 2020; Zhang, Wang, & Liang, 2021). All companies face competition, which drives them to innovate and gain top management support. The data suggest that government support is crucial for adopting digital transformation (H9). Public authorities' support encourages companies to go digital; this aligns with recent studies (Gui et al., 2020; El-Haddadeh, 2021). In the face of pressure from partners (H10), the findings indicate that pressure from business partners has a beneficial influence on organizations' adoption of digital transformation. The need for a relationship with business partners drives this, so we must consider this relationship in digital transformation, as prior work explains (Lin & Lin, 2008). The results indicate that the legal framework (H11) encourages companies to adopt digital transformation, consistent with previous studies (Maroufkhani et al., 2023). This strong relationship can be explained by the need for a legal framework and government regulations to engage in digital technologies.
The results confirm the idea that emerging environments do not always respond linearly to Western models. The absence of the effect of certain variables invites a rethinking of how digital maturity is built in a transitioning ecosystem; this represents an opportunity to reposition the TOE framework to better reflect African realities.
Our work does not seek to reinvent the model, but to question its explanatory scope in a context that is still little explored. The originality of this research thus lies in the empirical analysis of the limitations of the TOE model when applied in a developing environment, highlighting the factors that, contrary to theoretical expectations, do not show any significant effect (such as company size, resource availability or organizational readiness). These results make an important contribution to the literature, suggesting that certain theoretical levers are not universally applicable and that the adoption of digital technology in Morocco follows specific logics that warrant further exploration.
Conclusion
Today, companies worldwide face industrial changes and crises. Companies in Morocco are no exception. As they begin to introduce the concepts of digitalization into their practices, this work stands out for its originality; this study contributes to the limited body of empirical research examining the determinants of digital transformation adoption among Moroccan firms.
The elements that support Moroccan businesses' adoption of digital strategies are investigated in this study. Using the TOE framework, it looks into the causes of digital transformation in companies. The PLS-SEM approach was employed. The findings indicated that the adoption of digital transformation is influenced by eight factors. Among these are the following: organizational (top management support), technological (relative advantage, compatibility) and environmental (competitive pressure, government support, partner support and the legal framework). The findings demonstrated that four factors (digital employee, digital cost, firm size and organizational readiness) had no significant effect on the adoption of digital transformation by firms. On the other hand, top management's support for digital transformation is significantly impacted by relative advantage and competitive pressure.
First, the TOE research paradigm serves as the foundation for this work's findings, which examine the elements influencing the adoption of digital transformation. Future work can focus on other models outside the TOE framework, such as TAM and UTAUT. Second, this study did not specify company size or business sectors. It analyzed all Moroccan companies, regardless of size and sector. Thus, we must stress the need for similar studies. They should focus on the factors affecting the adoption of a digital strategy in a specific sector and company size.
The main objective of this theoretical framework is to justify why this study should not simply be considered an application of the TOE, but rather an authentic and contextual assessment of the theory's applicability to an emerging economy, which is the framework's primary contribution to scholarship and practice.
Nevertheless, we recognize that using a single method, in this case a quantitative approach, is a limitation in and of itself. This is one of the reasons we suggest in the research agenda the need to extend this research to mixed method designs that incorporate quantitative and qualitative components to further explore the construct of digital adoption.
The findings of this research offer a number of valuable insights. For public policymakers, the findings suggest specific actions, such as enhancing support structures, fostering innovation financing and developing digital skill training aligned with market demand.
On a theoretical level, this research contributes to the literature by demonstrating that the TOE model is not universal. It reveals the limitations of certain organizational and environmental dimensions (such as company size, digital costs and level of preparedness) in the Moroccan context. More specifically, our findings refine the TOE framework by identifying clear boundary conditions in which certain organizational and environmental variables traditionally considered significant – such as firm size, digital cost and organizational readiness – lose their explanatory power in an emerging economy context. This suggests that the TOE model should not be applied as a universal framework, but rather adapted to institutional, economic and organizational specificities.
The study contributes to previous studies in assessing the degree to which the digital adoption drivers are context-specific, negating the socio-economic context and the emerging economies applicability of the proposed models in conjunction with the core theories. This stance advocates for the positioning of the leadership construct in the TOE framework, whilst also underscoring the institutional specificities where the TOE model fails. This is the case for Morocco, which, as a result of its dominant SMEs, is structurally similar to a number of other emerging economies and ongoing digital policy reforms.
From a managerial perspective, the results indicate that, for Moroccan firms, digital transformation is mostly the result of strategic and leadership factors, as opposed to the more traditional structural factors like firm size and digital costs. Top management support is particularly important, reinforcing the idea that digital transformation initiatives should be considered as high-priority strategic initiatives, and should not be viewed as purely technological investments. Additionally, the notable importance of the factors relative advantage and compatibility suggests that managers ought to focus their attention on digital solutions that more closely conform to the organizational frameworks and that provide value in terms of improved operational efficiencies. In addition, the competitive and partner pressures described above indicate the importance of organizational networks and environmental scanning on the adoption of digital technologies. Overall, the findings indicate that even with limited resources, digital transformation is feasible for firms that possess strong leadership, a cohesive strategy and external support.
There are several limitations in this study. The first limitation stems from the focus on Moroccan companies, which makes the findings less generalizable beyond this context. The second limitation is that the cross-sectional design of the study does not capture any of the changes that happen to digital transformation over time. Finally, although effective, the TOE model has limitations in explaining the structural and institutional specificities of Moroccan SMEs, suggesting the need for an enriched or hybrid model in future studies. These limitations open up prospects for enriching the TOE model through a contextualized reformulation adapted to emerging economies.
Future research could take a more nuanced approach to the internal particularities of firms, especially their dynamic capabilities and human capital, through a blended framework of TOE and resource-based theory.
Also, multi-period studies will facilitate the assessment of changes and impacts of digitalization alongside the public sector reforms and the Morocco Digital Horizon 2030 national strategy. Such a lens will provide further richness and granularity to the understanding of the digital transformation phenomenon in Morocco.
The supplementary material for this article can be found online.
