The Technology, Organization, and Environment Theory is an organizational-level approach originally proposed by Tornatzky and Fleischer (1990) used to understand the role of three crucial components within a company that can impact decisions regarding the implementation of new technologies (Souza et al., 2017; Matikiti et al., 2018). These are the technological, organizational, and environmental contexts (Eze et al., 2021; Matikiti et al., 2018; Picoto et al., 2021).

This theory is one of the most commonly used models to explain the adoption and implementation of Internet technologies (Gonçalves et al., 2016; Matikiti et al., 2018; Melo et al., 2021), focusing on studying adoption within companies where each individual plays a different role, necessitating adaptation both within the organization and regarding the innovation itself (Souza et al., 2017). In the context of SMEs, many adoption models are centered on the entrepreneur's perception, attitude, or beliefs and are often user-centric. However, it is necessary to consider organizational factors and the external environment (Souza et al., 2017; Eze et al., 2021).

The technological context refers to internal and external technologies relevant to the company. It can also denote the necessary skills required to use that specific technology. The organizational context includes the company's resources, the linkages between employees, company size, and available resources. The environmental context encompasses external factors influencing a company's decision to adopt new technologies, including competitors, customers, and government involvement (Matikiti et al., 2018; Picoto et al., 2021).

Thus, authors Eze et al. (2021) highlight that the TOE theory provides a framework for assessing the adoption of SMM; as it emphasizes both internal aspects (e.g., technical knowledge and cost reduction) and external aspects (e.g., competitive pressure and customer pressure) of an organization that can influence the adoption of new technologies. In this way, the theory has proven useful in analyzing technology adoption by companies. And in the case of SMM adoption by micro and small enterprises, it proves to be pertinent.

SMM uses technologies, social media platforms, channels, and software to create, communicate, and deliver value-laden exchange offerings for the company and customers (Yao et al., 2019). From a marketing perspective, SMM is reshaping traditional marketing methods that involve a one-way flow of information and high production and distribution costs (Tarsakoo & Charoensukmongkol, 2019).

It is understood that whenever an individual perceives that using a particular technology will enhance performance or benefit the company, they will be eager to adopt it. Likewise, if the new technology is considered easy to implement, people will not hesitate to adopt it. Therefore, if SMEs comprehend the advantages of utilizing Social Media Marketing and can use it without challenges, they can more readily accept the adoption and use of Social Media Marketing (Almeida et al., 2017; Matikiti et al., 2018).

It is observed that the use of SMM does not require significant investments and, if used effectively, can provide opportunities for micro and small enterprises to access a large customer base (Ashish Kumar & Möller, 2018; Silva et al., 2020; Tarsakoo & Charoensukmongkol, 2019). Furthermore, it enables customers to directly connect with new products, services, and brands easily, making micro and small businesses more competitive (Pandey et al., 2020), more profitable, and increasing the company's overall value (Patma et al., 2020).

However, in the context of SMEs, which is still underexplored, SMM needs to consider various factors contributing to or inhibiting their adoption. Despite the advantages of SMM adoption, company size, the manager's age, and the extent of the company's innovation capacity are essential factors and positively affect the likelihood of adoption (Gutierrez-Leefmans et al., 2019). Furthermore, Yao et al. (2019) corroborate that geographical and demographic characteristics of companies have a significant effect on adoption.

Additionally, according to Chatterjee and Kar (2020), SMM is a viable tool that can help businesses attract customers; nevertheless, the adoption rates of SMM by SMEs have remained low due to factors that inhibit or hinder adoption. They mention that a lack of knowledge about utilizing the technology's benefits is a significant barrier to adoption. Alford and Page (2015) highlight that factors such as cost, lack of time, and difficulty in integrating information can inhibit the adoption of such technologies.

Thus, this study examines constructs' roles concerning perceived benefits when adopting and using SMM.

2.2.4 Perceived Benefits (PB)

The primary objective of companies is to maximize profits and financial gains (Patma et al., 2020). The study by Matikiti et al. (2018b) highlights that if a company believes that SMM possesses desirable attributes that can enhance performance, it is more likely to adopt and use it. Moreover, benefits have been considered the most prominent external factors influencing the use of social media marketing.

In this regard, the work of Patma et al. (2020) yields one of its results indicating that perceived benefits positively affect the adoption of the tool. Thus, it is understood that the adoption of SMM will be enhanced if entrepreneurs can identify and perceive the benefits achieved through this strategic tool. This study proposes that:

H5: The perception of benefits is positively related to the level of tool usage.

As seen in Figure 1, in light of the literature, the argument put forth here is that the constructs "technical knowledge," "cost reduction," "customer pressure," and "competitive pressure" play a role as antecedents to the attitude of adoption, encompassing technological, organizational, and environmental aspects, which in turn relate to the perceived benefits of using SMM.

Figure 1.
Proposed Conceptual Model.
Source: Author's own elaboration.

The research instrument consisted of a demographic data section: gender, age, and level of education, as well as information regarding sectorization, years of existence, number of employees, and technical knowledge, for sample characterization purposes.

The questionnaire contained measurement scales for each construct: technical knowledge (TK), cost reduction (CUS), customer pressure (CUP), competitive pressure (COP), perceived benefits (PB), and level of usage (LU), as presented in Table 1. The scales were 7-point Likert-type scales, ranging from (1) strongly disagree to (7) strongly agree with the statements.

The second part of the questionnaire comprised questions aimed at gathering information about the entrepreneurs' characteristics and their businesses. A pre-test of the instrument was conducted to enhance the questionnaire's comprehensibility from the respondents' perspective. In the pilot study using this pre-test, 20 respondents meeting the criteria for the target audience of the main sample were obtained. After considering their feedback, some changes were made to the final questionnaires.

Table 1
Measurement Scale of the Proposed Constructs

Source: Author's elaboration.

The study focused on the process of adopting SMM. In other words, the relationships between variables (model structure) according to the theoretical foundation (literature review) that supported the construction of hypotheses. Based on this, the sample size required for this study was calculated based on Hair et al.'s (2018) recommendation for a desired level of 15-20 observations per parameter in Structural Equation Modeling (SEM). The initial sample consisted of 338 respondents with six constructs measured by 19 psychometric items deemed sufficient (338 > 15*19 = 285) and well above the recommended threshold.

Therefore, to reach this desired research audience, a non-probability and convenience sampling method was adopted, as Malhotra (2019) suggested, which relied on clients' availability and interest to participate in the research. The online questionnaires were sent to entrepreneurs via WhatsApp, and the assistance of Local Innovation Agents duly registered by Sebrae, who supported businesses with this profile, was also utilized.

Before the actual analysis, data screening was conducted (Tukey, 1977). The aim was to identify inconsistencies among responses. Once this process was completed, the database contained 338 valid observations (Table 2).

Next, the subsequent step was to assess the reliability and validity of the scales employed. The following indicators were analyzed: Cronbach's alpha coefficient (CA), composite reliability (CR), average variance extracted (AVE), and convergent and discriminant validity (Hair et al., 2018; Fornell & Larcker, 1981; Cronbach, 1951).

The study used covariance-based Structural Equation Modeling (SEM) for data analysis. Following Anderson and Gerbing's (1988) approach, this statistical procedure involves two stages. First, a measurement model is developed to assess model fit. Then, structural modeling is used to determine the best-fitting model for testing causal relationships between independent and dependent variables. Statistical software R, version 4.2.0 for Windows, was employed to conduct the analysis.

A total of 338 respondents' data were collected, with the majority being female (53.85%) and having completed a bachelor's or postgraduate degree (58.28%). All interviewees used some form of social media in their businesses, with the primary platforms being Instagram and WhatsApp (95.86%), used on average 5.22 times per week with a median of 4.81, indicating active engagement by the respondents.

The mean age among the interviewees is 38.06 years, with a median of 37.51, and the mean business existence duration is 10.81 years, with a median of 7. Overall, the entrepreneurs are mature and experienced, yet this cannot be considered a criterion directly affecting tool usage. Furthermore, the average number of employees in the companies is 6.39, with a median of 4.39, signifying that most entrepreneurs have employees contributing to business growth. These data are presented in Table 2.

Table 2
Sample Characterization

Source: Author’s elaboration.

Thus, the sample proved suitable for the study. Most respondents utilize only Instagram and WhatsApp, indicating that these two social media platforms primarily drive the adoption of Social Media Marketing. At the same time, Facebook, Google, LinkedIn, Pinterest, TikTok, and YouTube do not hold significant relevance for entrepreneurs, collectively representing 4.16%.

Furthermore, despite most entrepreneurs possessing a high level of education, this does not necessarily imply adequate technical knowledge for tool utilization. Only 5.62% of entrepreneurs have incomplete secondary education, and 36.09% have completed secondary education or have incomplete higher education, categorizing them as having intermediate education levels.

Considering the criterion for convergent validity analysis set by Fornell and Larcker (1981), the average variance extracted (AVE) values should be greater than 0.50. Thus, based on the information in Table 3, a satisfactory result was obtained for the analyzed model.

Regarding the reliability and internal consistency of the model, the values of Cronbach's Alpha demonstrated satisfactory values (all values are > 0.7), confirming the internal validity of the utilized scale. On the other hand, composite reliability (CR) assesses whether the indicators of each construct truly represent them (Bagozzi & Yi, 1988), and values between 0.7 and 0.9 are considered good indicators.

The assessment of collinearity involved calculating the Variance Inflation Factor (VIF) for each construct (TK = 1.34, COS = 1.44, PC = 1.49, PC = 1.18, PB = 1.71, LU = 2.02). All indicators met the collinearity criteria with VIF values below five (Hair et al., 2018).

Thus, all indicators displayed values that meet the established criteria, demonstrating the construct validity, as shown in Table 3.

Table 3
Measurement Model: Reliability and Validity

Source: Author’s elaboration.

The items PCL3 and PCO3 were excluded as their factor loadings were low and did not contribute to the formation of the construct.

Considering the Fornell & Larcker (1981) criterion, discriminant validity was assessed, as the square roots of the Average Variance Extracted (AVE) values for each construct are more significant than the correlations between the constructs, as can be observed in Table 4:

Table 4
Discriminant validity and means and standard deviations of responses on the constructs

Source: Author’s elaboration.

To assess the fit of the proposed model to the collected data, the method proposed by Byrne (2013) was applied to analyze fit indices for evaluating the adequacy of the proposed model to the sample data.

Thus, various indicators were used for estimation analysis and fit quality evaluation: χ² (chi-square), Df (degrees of freedom), χ²/Df (chi-square divided by degrees of freedom), NFI (normed fit index), IFI (incremental fit index), TLI (Tucker-Lewis index), CFI (comparative fit index), GFI (goodness-of-fit index), RFI (relative fit index), RMSEA (root mean square error of approximation), and SRMR (standardized root mean square residual).

Thus, various indicators were used for estimation analysis and fit quality evaluation: χ² (chi-square), Df (degrees of freedom), χ²/Df (chi-square divided by degrees of freedom), NFI (normed fit index), IFI (incremental fit index), TLI (Tucker-Lewis index), CFI (comparative fit index), GFI (goodness-of-fit index), RFI (relative fit index), RMSEA (root mean square error of approximation), and SRMR (standardized root mean square residual).

GFI and RFI should have values between 0 and 1, which allows GFI to represent the covariance between observed and model-calculated variables. For this, the indices need to exceed the value of 0.90 (Yilmaz, 2018). Therefore, the model met the criteria and is considered a good fit (GFI = 0.912 and RFI = 0.910). As for the population approximation error, RMSEA, the obtained value of 0.083 is considered acceptable. While the lower the SRMR value, the better the model fit (Yilmaz, 2018). Thus, the model indicates a good fit, as the SRMR value is 0.048, which is acceptable.

All indices were found to be appropriate, indicating model fit, see Table 5. Thus, the values yielded satisfactory results, demonstrating the adequacy of the proposed model for the collected data. Subsequently, the analysis proceeded to the structural model, namely, hypothesis verification.

Table 5
Model Fit Indexes

Source: Author’s elaboration.

As Anderson and Gerbing (1988) suggested, the research hypotheses are tested in this stage. Thus, technical knowledge and cost reduction were significantly related to perceived benefits (β= 0.167, t= 5.568, p < 0.05; β= 0.091, t= 3.425, p < 0.05), supporting hypotheses H1, H2.

Customer pressure and competitive pressure are also related to perceived benefits (β= 0.271, t= 6.175, p < 0.05; β= 0.068, t= 2.230, p < 0.05), supporting hypotheses H3, H4. Table 6 presents the results concerning the postulated hypotheses.

Table 6
Hypotheses of the Final Model

Source: Author’s elaboration.

It can be observed that technical knowledge and customer pressure are the constructs with the most vital relationship with perceived benefits, obtaining the highest beta values (0.167 and 0.271). However, competitive pressure has a weaker influence on the relationship with perceived benefits.

Furthermore, another analysis concerns the evaluation of Pearson's coefficient of determination (R²), which indicates the percentage of variation in the dependent variable explained by the independent variables. According to Cohen (1988), R² is considered a small effect when the value is 2%, a medium effect when the value is 13%, and a significant effect when R² is equal to or greater than 26%.

It was observed that technical knowledge, cost reduction, customer pressure, and competitive pressure explain about 88% (p-value < 0.001) of the perceived benefits by entrepreneurs. In addition, perceived benefits explain about 44% (p-value < 0.001) of the Social Media Marketing usage level in micro and small enterprises, indicating a large effect.

In line with the literature, the research highlighted that the Technology-Organization-Environment (TOE) model identifies three contexts that influence the process through which an organization adopts and implements the SMM (Social Media Marketing) technological tool: technological, organizational, and environmental. Based on these contexts, variables were selected to analyze the adoption process of Social Media Marketing by micro and small enterprises in Paraíba, considering certain conditioning factors and inherent barriers.

The findings of this study demonstrate that the entrepreneur's technical knowledge is positively related to the perceived benefits of adopting Social Media Marketing. This implies that the greater the entrepreneur's technical knowledge, the higher the perception of the benefits of utilizing SMM as a strategic tool. This finding aligns with the outcomes from the studies of Dahnil et al. (2014), Patma et al. (2020), and Tarsakoo and Charoensukmongkol (2019). This suggests that higher technical knowledge leads to a heightened perception of the benefits associated with tool usage. Specifically, regarding micro and small enterprises (MSEs), inadequate technical knowledge could hinder the effective implementation of SMM, highlighting it as a significant barrier.

An essential aspect of SMM adoption is costs (Chatterjee & Kumar, 2020). Kumar and Möller (2018) add that traditional media usage in marketing activities is expensive, so companies increasingly embrace SMM for its cost-effective solution. Furthermore, companies would not adopt SMM if the initial adoption expenses were high (Yao et al., 2019).

In this sense, the current research corroborates the literature. This suggests that cost reduction plays a conditioning role in using SMM by MSEs, meaning that cost reduction is positively related to the perceived benefits of adopting Social Media Marketing. This implies that the higher the perception of low costs, the greater the perception of benefits in utilizing SMM as a strategic tool.

Furthermore, this study revealed that customer pressure positively relates to the perceived benefits of adopting Social Media Marketing. In other words, the greater the customer pressure, the higher the perception of the benefits of using SMM as a strategic tool. This is consistent with other studies: Patma et al. (2020) demonstrated that external pressure positively affects SMM adoption. However, Matikiti et al. (2018) couldn't establish a relationship between customer pressure and attitude toward SMM adoption; although it is not the same related variable, it holds significance in the theme.

Within external pressure, beyond customer pressure, the pressure from competitors is also included (Dahnil et al., 2014). This study further confirmed that competitive pressure positively relates to the perceived benefits of adopting Social Media Marketing. Thus, it can be asserted that the greater the pressure from competitors, the higher the perception of the benefits of utilizing SMM as a strategic tool.

Based on this, Patma et al. (2020) concluded that due to external pressure on MSEs, some competitors have adopted SMM in their companies to reach new customers, greater profitability, and competitive advantage. This is because competitors recognize the importance of SMM and use it in their business activities. They also perceive that MSE customers are ready to conduct business online through social media applications and are becoming increasingly demanding and quick in their interactions and purchasing processes.

The results also demonstrate that perceived benefits influence entrepreneurs' adoption and use of SMM. Suppose individuals perceive that using SMM will benefit or provide a competitive advantage to their business. In that case, their attitude toward social media adoption will be positive, as well as their continued usage.

According to researchers Praveena and Thomas (2014) and Shen (2015), if an organization has a positive attitude toward SMM adoption, it will want to continue using it.