Maturity in Indicator Management and the Effectiveness of Data-driven Decisions in Brazilian Companies

Aser Gomes; Fabiano Sannino

DOI: 10.22167/2675-6528-202602692

Article derived from a Course Conclusion Work (TCC), with content based on the student’s original work and adapted to the editorial format of the E&S Magazine with the support of the ResumeAI tool, an artificial intelligence solution developed by Instituto Pecege for textual synthesis and organization.

Summary

A study investigated the relationship between KPI governance, data-driven culture, technical capabilities, and decision-making effectiveness in Brazilian organizations. The objective was to identify analytical maturity profiles and analyze the correlation between these dimensions and decision-making effectiveness. A survey was conducted with 42 respondents, using a 20-item Likert questionnaire. Descriptive analyses, Spearman correlation, multiple regression, k-medoids cluster analysis, and exploratory factor analysis (EFA) were applied, with multicollinearity and common method bias tests. Maturity dimensions strongly correlated with decision-making effectiveness. The regression explained 79.5% of the variance, with analytical capabilities and data-driven culture as significant predictors. KPI governance did not show a direct effect due to multicollinearity. EFA extracted three factors, which diverged from the original theoretical model. Cluster analysis identified Beginner, Intermediate, and Advanced profiles, demonstrating that advanced organizations achieved 2.02 times higher effectiveness. Harman’s test indicated moderate common method bias. It was concluded that analytical capabilities and data-driven culture were the central determinants of decision-making effectiveness, highlighting a mismatch between formal and behavioral aspects. Limitations included the convenience sample, the cross-sectional design, and the moderate common method bias.

Keywords: Exploratory factor analysis; Analytical capabilities; Data-driven culture; Performance indicator governance; Analytical maturity.

1. Introduction

The acceleration of digital transformation, driven by the emergence of generative Artificial Intelligence tools, has redefined the competitive paradigms of organizations. In the global scenario, the so-called fourth industrial revolution has imposed on companies the need to reconfigure business models, decision-making processes, and governance structures, with data assuming centrality (Schwab, 2016).

In the Brazilian market, this transition presented specific contours. While the adoption of advanced technologies, such as generative Artificial Intelligence, is reported by a significant portion of companies, the centralized and strategic integration of these technologies remains a challenge. Historical research already indicated considerable heterogeneity between sectors and organizational sizes in the ability to convert technology investments into management results (Bain & Company, 2024; Diniz et al., 2009).

In this context, the management of Key Performance Indicators (KPIs) has emerged as a fundamental pillar of organizational transformation. KPIs represent the systematic process of defining, monitoring, and analyzing metrics that translate strategic objectives into measurable results, functioning as operational translations of strategy (Fleury and Fleury, 2001). Systems such as the Balanced Scorecard, for example, have operationalized this alignment, distributing strategic objectives into indicators across various organizational perspectives (Kaplan and Norton, 1997).

Data-driven decision-making, in turn, constitutes the most advanced stage of organizational decision-making, where strategic and operational choices are based on systematic analysis of evidence, aiming to minimize cognitive biases and uncertainties. This approach transcends the mere use of technological tools, configuring itself as a multidimensional organizational culture (Pereira, 2023), aligned with integrated approaches to process management and information technology (Laurindo and Rotondaro, 2006).

However, the mature implementation of these systems in Brazil faced particular challenges. Empirical research in Brazilian industrial companies demonstrated that, although the formal adoption of indicator systems was associated with greater strategic clarity and better operational results, implementation maturity varied significantly across sectors (Hourneaux Junior et al., 2005). National literature had already pointed out, since the late 1990s, structural difficulties in integrating the strategic, tactical, and operational levels of performance measurement systems (Martins and Salerno, 1998). Furthermore, the regulatory framework, including the General Data Protection Law (LGPD), required KPI governance to incorporate dimensions of legal and ethical compliance in information processing (Bioni, 2021; Hirano, 2024).

The theoretical premise that KPI governance, data-driven culture, and technical capabilities synergistically converge to enhance decision-making is widely recognized in the analytics management literature (Davenport and Harris, 2007; Hirano, 2024). However, the persistence of “structural formalism” and behavioral gaps in Brazilian organizations, such as low autonomy and aversion to error, suggest a mismatch between the existence of formal systems and their effective appropriation by the team (Caldas and Wood Jr., 1999; Barros and Prates, 1996). This practical and theoretical gap in the national context justifies the need to investigate how these elements relate and influence decision-making effectiveness. Given this context, the general objective of the present work was to investigate the correlation between maturity in KPI management and the perception of effectiveness of data-driven decision-making in companies operating in Brazil.

2. Material and Methods

The research adopted a quantitative, descriptive, and correlational approach, using a cross-sectional survey design. The methodological objective was to test hypothetical relationships between theoretical constructs, following a deductive approach based on the revised framework. The study involved data collection with human participants through a structured questionnaire, with free and informed consent. The Ethical Considerations Form was validated by the advisor, and the project, dealing with professional perceptions without sensitive data, did not require submission to a Research Ethics Committee. The data were treated in an aggregated manner, ensuring participant anonymity.

The data collection instrument was developed based on the literature of analytical maturity (Pereira, 2023) and performance management (Fleury and Fleury, 2001; Kaplan and Norton, 1997). The constructs were operationalized into measurable items, decomposing maturity in KPI management into three dimensions: governance, culture, and technical capabilities. The operationalization followed content validation (Sordi and Meireles, 2011), with constructs adapted from the literature, without direct derivation from instruments validated by Confirmatory Factor Analysis. The final questionnaire was structured into six sections, including demographic data, 20 items on maturity dimensions (KPI Governance, Culture, Technical Capabilities), and four items for Decision Effectiveness. All maturity and effectiveness items used a five-point Likert scale (1 = Totally Disagree; 5 = Totally Agree), with a “Not Applicable” option to reduce forced response bias (Bortolotti et al., 2012).

Data collection occurred between March 18 and 30, 2026, through dissemination on LinkedIn and the researcher’s professional contacts. 44 records were obtained, of which 42 were considered valid after excluding cases that presented “Not Applicable” responses in four or more structural items. Initially, the research was planned to cover analytics and management professionals, but it was expanded after March 26 to non-specialized groups due to low adherence, which introduced intentional heterogeneity in the respondents’ profile.

The data were processed and analyzed following a structured protocol. “Not Applicable” responses were recoded as nulls, and records with four or more missing items were discarded. Composite scores were calculated for each construct (arithmetic mean) and outliers were checked. Descriptive analysis included basic statistics and Cronbach’s alpha coefficient for internal reliability (Hair et al., 2009; Field, 2009). For scores and correlations, the pairwise deletion method was adopted, which maximizes data utilization in the presence of sparse missing values. In inferential analysis, Spearman Correlation (Field, 2009), Multiple Linear Regression (Hair et al., 2009), and k-medoids Cluster Analysis (Partitioning Around Medoids – PAM) (Kaufman and Rousseeuw, 1990), robust for ordinal data, were employed.

To control for common method bias (CMB), recommendations from Podsakoff et al. (2003) were followed, including anonymity, information about the absence of right or wrong answers, a “Not Applicable” option, and structural separation of predictor and criterion sections. As a post hoc statistical procedure, Harman’s Single-Factor Test was applied, subjecting the 20 items to a Principal Component Analysis without rotation. The first factor explained 58.76% of the total variance, indicating moderate CMB. An Exploratory Factor Analysis (EFA) was also conducted with PCA and promax rotation (Hair et al., 2009; Field, 2009), assessing the adequacy of the correlation matrix using Bartlett’s and KMO tests (Malhotra, 2019). The EFA indicated a three-factor solution, explaining 61.6% of the total variance after rotation. The analyses were processed in the Google Colab environment with the `factor_analyzer` library.

The sample sufficiency for multiple linear regression with three predictors was evaluated by Cohen’s method (1988), which considers the expected effect size, significance level, and desired statistical power. The observed R² of 0.795 (f² = 3.878), classified as a very large effect, indicated that the sample of 42 respondents was more than sufficient to detect the effect found, given a minimum n of seven for a power of 0.80.

3. Results and Discussion

This section details the research findings, interpreting them in light of the research problem and the adopted theoretical framework. Initially, the sample characterization is presented, followed by descriptive statistics of the analytical maturity and decision-making effectiveness dimensions. Next, the correlation, multiple linear regression, cluster, and exploratory factor analyses are discussed, which allowed for hypothesis testing and pattern identification. Finally, the managerial implications and strategic propositions for each maturity profile are addressed, culminating in an interpretive synthesis of the results and their contributions to understanding analytical maturity in the Brazilian context.

Sample Characterization

The valid sample of the study comprised 42 respondents, predominantly from large organizations, with 61.9% of participants working in companies with over 300 employees. Medium-sized companies (101 to 300 employees) represented 19.0%, while small companies (31 to 100 employees) and microenterprises (up to 30 employees) accounted for 14.3% and 4.8%, respectively. In terms of sector of activity, Services was the most representative, with 31% of respondents, followed by Technology and Industry, both with 14.3% each. Retail and Health contributed 9.5% each, and Fintechs, Logistics, and Education contributed 7.1% each. The majority of respondents had up to six years of experience in their role, with 42.9% having up to three years and 38.1% between three and six years. Only 19.0% had more than six years of experience. Regarding their function, 31.0% worked in operational positions not related to analytics/technology, and 21.4% in leadership positions not related to these areas, indicating a diversity of perspectives.

Descriptive Statistics and Reliability

Descriptive analyses of the dimensions revealed that Analytics Capabilities presented the highest average (3.03), followed by Decision-Making Effectiveness (3.22), KPI Governance (2.98), and Data-Driven Culture (2.86). Cronbach’s alpha values for all dimensions ranged between 0.817 and 0.908, indicating high internal reliability of the constructs, according to the criteria of Hair et al. (2009). Detailed analysis of governance items showed that formal processes for defining and reviewing KPIs (average of 3.15) and regular monitoring (average of 3.10) obtained the highest scores. In contrast, employees’ understanding of KPIs registered the lowest average (2.67), suggesting a gap between the formalization of systems and effective team appropriation, a finding that resonates with the literature on behavioral barriers in the Brazilian context (Caldas and Wood Jr., 1999; Barros and Prates, 1996).

The decision-making effectiveness dimension had its improvement item in operational performance with the highest average (3.40), while the investment in analytical training (CULT5) obtained the lowest overall average (2.51). This comparison suggests that, although KPI management is associated with performance improvements, investment in developing employees’ analytical skills is still a weak point. The proximity between the averages and medians in all dimensions indicated a relatively symmetric distribution of responses, with a slight positive skew in Capabilities, which may reflect a slightly more optimistic perception or greater variability in the evaluations of this dimension.

Correlation Analysis

The three research hypotheses were confirmed, as all maturity dimensions (KPI Governance, Data-Driven Culture, and Analytics Capabilities) showed positive and strong correlations with Decision Effectiveness, with Spearman coefficients above 0.6. This result is in line with theoretical models that conceive analytical maturity as a multidimensional phenomenon, where the synergy between these elements is crucial for improving decision-making (Pereira, 2023; Hirano, 2024). The lowest correlation was observed between Governance and Effectiveness (0.738), which may indicate a limitation of isolated formalism, corroborating the Bain & Company report (2024) on the excess of indicators without effective action in Brazil. On the other hand, the strongest correlation, between Capabilities and Effectiveness (0.874), reinforces the idea that analytical capability is a fundamental prerequisite for data-driven competitiveness (Davenport and Harris, 2007; Hirano, 2024).

Multiple Linear Regression Analysis

The multiple linear regression model demonstrated excellent predictive power, with an adjusted R² of 0.779, indicating that the dimensions of Governance, Culture, and Capabilities jointly explained approximately 79.5% of the variance in Decision Effectiveness. The diagnostic analysis revealed severe multicollinearity, with an average VIF greater than 5, stemming from high correlations between the independent variables (p between 0.809 and 0.866). Although this condition is expected in theoretically interrelated constructs (Hair et al., 2009; Field, 2009), it inflates the standard errors of the coefficients, requiring caution in interpreting individual effects. Adopting a significance level of 5% (p < 0.05), Analytics Capabilities (β = 0.515; p < 0.001) and Data-Driven Culture (β = 0.408; p = 0.017) were statistically significant predictors of Decision Effectiveness.

In contrast, KPI Governance (β = 0.012; p = 0.936) did not achieve direct statistical significance in the regression model. This result suggests that, in the analyzed context, investment in technological infrastructure, professional expertise, and data governance (Capabilities) were the most robust predictors, followed by a data-driven decision-making culture. The lack of direct significance of Governance, despite its correlation with Effectiveness (0.738), can be interpreted as an effect mediated by the other dimensions. In other words, formal governance seems to translate into effectiveness only when accompanied by an effective analytical culture and capabilities, as pointed out by Caldas and Wood Jr. (1999) and Barros and Prates (1996) regarding the need for alignment between indicators, competencies, and organizational behavior (Kaplan and Norton, 1997; Laurindo and Rotondaro, 2006).

Cluster Analysis

K-medoids clustering identified three organizational maturity profiles, with statistically significant differences between them (p < 0.05). The Beginner profile, which grouped 19 organizations (45% of the sample), showed averages below 2.10 in all dimensions and a decision-making effectiveness of 2.30. The Intermediate profile, with 13 organizations (31%), registered averages between 2.95 and 3.35, and an effectiveness of 3.35. Finally, the Advanced profile, composed of 10 organizations (24%), exhibited averages above 4.10 in all dimensions and a decision-making effectiveness of 4.65. Organizations in the Advanced profile achieved a decision-making effectiveness 2.02 times higher than those in the Beginner profile (4.65 versus 2.30), corroborating findings in the literature on analytical capability as a source of competitive advantage (Davenport and Harris, 2007).

The predominance of the Beginner profile (45%) and the smaller proportion of the Advanced profile (24%) indicate that the majority of the surveyed organizations are still in the early stages of their analytical journey. Notable gaps were observed in cultural aspects, such as low investment in analytical training (average of 2.51) and lack of autonomy for KPI adjustments (average of 2.98). These results are consistent with the behavioral barriers to data-driven management in Brazil, as discussed by Caldas and Wood Jr. (1999) and Barros and Prates (1996). The Bain & Company report (2024) also corroborates this scenario, pointing out that, despite the existence of analytical infrastructure, deficiencies in human competencies and the culture of experimentation persist.

Common Method Bias Control

Given the cross-sectional design and single-source self-report data collection, the research was susceptible to common method bias (CMB). To mitigate its effects, procedural controls recommended by Podsakoff et al. (2003) were implemented, including ensuring anonymity and confidentiality, explicitly informing participants that there were no right or wrong answers, and including a “Not Applicable” option for Likert items. Structural separation of questionnaire sections, positioning predictor constructs before the criterion variable, also served as a partial control. Harman’s Single-Factor Test, applied as a post hoc statistical procedure, revealed that the first unrotated factor explained 58.76% of the total variance of the 20 items. This value, exceeding the recommended 50% cutoff, indicated a moderate threat of CMB, suggesting that the bias may have inflated correlations and regression coefficients. However, the literature acknowledges that Harman’s Test is conservative and can be influenced by high theoretical correlation between interdependent constructs, which is expected in organizational maturity models (Richardson et al., 2009).

Sample Sufficiency and Statistical Power

The sample size sufficiency evaluation for multiple linear regression with three predictors, using Cohen’s (1988) method, indicated that for the observed R² of 0.795 (f² = 3.878), a considered very large effect, the minimum number of respondents for a statistical power of 0.80 would be only seven. With 42 respondents, the sample proved more than sufficient to detect the effect size effectively found in the data. However, this evaluation should be interpreted with caution, as the observed f² is exceptionally high, possibly inflated by the diagnosed CMV and the low variability of the Likert scale. If the actual effect were of moderate magnitude (f² = 0.15), the sample of 42 respondents would be insufficient, requiring a minimum of 68, according to the recommendations of Hair et al. (2009) for medium effects with an alpha of 0.05 and a power of 0.80.

Exploratory Factor Analysis

To verify the dimensional structure of the instrument, an Exploratory Factor Analysis (EFA) was conducted based on the 20 scale items, using principal component extraction and promax oblique rotation. The adequacy of the correlation matrix was confirmed by Bartlett’s test of sphericity (χ²(190) = 757.96; p < 0.001) and the Kaiser-Meyer-Olkin (KMO = 0.848) sampling adequacy measure, classified as excellent (Malhotra, 2019). Analysis of the eigenvalues, following Kaiser’s criterion, indicated the retention of three factors, which cumulatively explained 61.6% of the total variance after rotation. Factor 1, named Analytical Infrastructure and Autonomy, explained 19.1% of the variance and grouped items related to technological capabilities (infrastructure, data governance), autonomy for KPI adjustments, and decision-making agility, suggesting a dimension focused on technical resources and freedom of action. Factor 2, Formalization and Sharing, explained 17.7% of the variance, grouping items on formal KPI management processes (definition, monitoring), understanding of indicators by collaborators, and sharing of insights, reflecting the structural and collective nature of performance management. Factor 3, Effectiveness and Learning Culture, explained 24.8% of the variance, capturing items on results (ROI, uncertainty reduction, operational improvement) combined with data-driven leadership, experimentation, and continuous investments, indicating the integration between analytical culture and perceived effectiveness.

The correlations between the factors ranged from 0.544 to 0.646, confirming the theoretical interdependence of the constructs and justifying the choice of oblique rotation (Hair et al., 2009). The identified three-dimensional solution partially diverged from the original four-dimensional theoretical model (governance, culture, capabilities, and effectiveness), but was consistent with the high multicollinearity diagnosed in the regression and with the literature pointing to the empirical overlap between governance, culture, and capabilities in the early stages of analytical maturity (Davenport and Harris, 2007; Hirano, 2024). This divergence between the theoretical model and the empirical factorial solution represents an important limitation, suggesting that respondents may not have clearly distinguished the four proposed conceptual dimensions, reinforcing the need for future psychometric validations of the instrument.

Managerial Implications and Strategic Propositions by Maturity Profile

Beginner Profile

The Beginner profile, representing 45% of the sample, is characterized by averages below 2.10 in all dimensions and a decision-making effectiveness of 2.30, indicating organizations that have not yet structured their analytical journey. Its strengths include a “strategic tabula rasa,” which allows the adoption of best practices without the cost of reengineering legacy systems, and low organizational complexity, which favors the implementation of cultural changes with less bureaucratic inertia (Barros & Prates, 1996). Weaknesses are critical infrastructure scarcity, the absence of formal governance, and a culture of error aversion, which inhibits experimentation. Opportunities lie in the appropriation of open-source tools, which can generate quick visibility gains with low investment, and in basic data literacy training, which can enhance employees’ understanding of data value. Threats include leadership skepticism, which can render analytical initiatives unfeasible, and the “eternal pilot syndrome,” where pilot projects do not scale due to lack of resources or governance (Bain & Company, 2024).

For this profile, the strategic proposition is the “Analytical Foundation”, focused on establishing an analytical minimum viable product (MVP). It is recommended to define a basic set of up to five critical KPIs, accompanied by a simple executive dashboard. Simultaneously, a “data champions” program should be implemented to disseminate the experimentation culture within teams. The initial success metric should not be accounting ROI, but rather the reduction of uncertainties and the frequency of dashboard usage. The trap to be avoided is investing in complex technology before consolidating the governance discipline, which could aggravate the infrastructure shortage and the lack of formal governance.

Intermediate Profile

The Intermediate profile, covering 31% of the sample, represents organizations with reasonable infrastructure (Capabilities = 3.30) and formal processes (Governance = 3.20), but which have stagnated in their evolution, with averages between 2.95 and 3.35 and an effectiveness of 3.35. Culture (2.95) is the main bottleneck, especially the low investment in training (CULT5 = 2.51) and limited sharing of insights (CULT4). Their strengths include an operating infrastructure, with analytical tools that already meet basic needs, and structured monitoring processes that generate historical data. Weaknesses are a fragile learning culture, where errors are not seen as inputs for improvement, low autonomy for tactical KPI adjustments, and underinvestment in talent. Opportunities involve leveraging existing infrastructure to focus on data interpretation and action, and Data Literacy programs at scale for the entire organization. Threats are stagnation in mediocrity and turnover of dissatisfied talents due to lack of autonomy and learning.

The strategic proposition for this profile is “Behavioral Activation”, aiming to break data isolation. The strategy should focus on creating communities of practice to stimulate the sharing of insights (Nonaka & Takeuchi, 1995) and on implementing dedicated hours for experimentation, encouraging collaborators to test hypotheses without fear of failure. In governance, controlled decentralization is suggested, granting autonomy for tactical KPI adjustments, but maintaining central governance for strategic indicators. The key indicator for this profile is the ROI of analytics, which must be measured to justify investment in training and prevent the data area from being seen as a cost center, combating underinvestment in talent and a fragile learning culture.

Advanced Profile

The Advanced profile, comprising 24% of the sample, shows averages above 4.10 and a decision-making effectiveness of 4.65, representing the study’s benchmark. The synergy between governance, culture, and capabilities is evident, resulting in more agile decisions and better operational performance. Its strengths include high decision-making effectiveness, a data-driven leadership culture, where executives use data for strategic choices, and a robust infrastructure with mature governance. Weaknesses are possible complacency, which may reduce innovative momentum, and not-yet-ideal capacitation, indicating that people development has not kept pace with technological development. Opportunities include innovation with Generative AI, leveraging the clean and governed database, and the creation of sustainable competitive advantage, making analytical maturity a distinctive capability (Teece, 2007). Threats include talent loss to the market and procedural rigidity, which can stifle creativity and rapid adaptation to change.

For the Advanced profile, the strategic proposition is “Innovation and Market Leadership”, with the recommendation to migrate from data-driven to data-centric innovation. This implies transforming the data area into a center of excellence that anticipates business needs, investing in training programs focused on data storytelling and algorithmic ethics, and institutionalizing the ROI measurement of each analytical project, linking it to objective financial indicators. The great challenge is to balance consolidated governance with full autonomy, allowing teams to experiment freely within ethical and data security boundaries. The success metric here is the ability to generate new business or revenue models from data, not just optimize the existing one, overcoming possible complacency and still not ideal capabilities.

Integrated Synthesis and Cross-cutting Patterns

The analysis of the results revealed a consistent cross-sectional pattern across all maturity profiles: the cultural dimension systematically presented the lowest scores when compared to governance and capabilities at the same stage. In the Beginner profile, culture (2.05) was lower than governance (2.10); in the Intermediate, culture (2.95) lagged behind capabilities (3.30); and in the Advanced, culture (4.10) was the lowest point of the tripod. This finding corroborates the central thesis of Caldas and Wood Jr. (1999) and Barros and Prates (1996), as well as Schein (2010), that the bottleneck of analytical transformation is not primarily technological, but behavioral. The mere existence of formal systems and processes does not guarantee effectiveness if there is no environment that encourages curiosity, experimentation, and the sharing of insights.

Regardless of the profile, research suggests an integrated approach that combines continuous investments in infrastructure, data governance, and visualization tools with leadership development programs, reward mechanisms for sharing insights, and the creation of an “error budget” that encourages calculated experimentation. The transition from the Beginner to the Advanced profile depends less on the volume of technology investment and more on how leadership models data usage behavior in daily life. The strategic propositions outlined for each profile provide a contextualized starting point, which should be adapted to the sector and size realities of each organization, aiming to overcome the behavioral and cultural barriers identified as the main obstacles to data-driven decision-making effectiveness.

4. Conclusion

This study investigated the correlation between KPI management maturity and the perceived effectiveness of data-driven decision-making in companies operating in Brazil. It was found that the dimensions of KPI governance, data-driven culture, and analytics capabilities correlated positively and strongly with decision-making effectiveness. Multiple regression analysis indicated that analytics capabilities and data-driven culture were statistically significant predictors of decision-making effectiveness, jointly explaining approximately 79.5% of the variance. KPI governance, although correlated, did not show a significant direct effect in the regression model, suggesting that its influence may be mediated by the other dimensions. Cluster analysis identified three maturity profiles — Beginner, Intermediate, and Advanced — demonstrating that organizations in the advanced stage achieved a decision-making effectiveness 2.02 times higher than beginners. A relevant cross-cutting finding was the observation that the cultural dimension systematically obtained the lowest scores in all profiles, highlighting a mismatch between structural formalism and behavioral aspects in the analytical journey of Brazilian organizations.

The study’s main contribution lies in proposing a contextualized strategic roadmap for each maturity profile, aiming to overcome the behavioral and cultural barriers identified as the primary obstacles to data-driven decision-making effectiveness. Limitations included the convenience sample and the cross-sectional design, which prevent definitive causal inferences and generalization to small businesses or other sectors. Additionally, moderate common method bias and severe multicollinearity necessitated caution in interpreting individual regression coefficients, and exploratory factor analysis revealed a three-dimensional structure divergent from the original theoretical model, suggesting empirical overlap between constructs. For future studies, longitudinal research is recommended to examine causality, the use of stratified sampling, and the inclusion of objective dependent variables, in addition to applying Confirmatory Factor Analysis and Structural Equation Modeling to enhance the instrument’s psychometric validation and the understanding of mediating effects.

Bibliographic References

Bain & Company. 2024. State of Data 2023–2024: um raio-x dos profissionais de dados do Brasil. Disponível em: <https://www.bain.com/pt-br/insights/state-of-data-2023_profissionais_dados/>. Acesso em: 20 fev. 2026.

Barros, B.T.; Prates, M.A.S. 1996. O Estilo Brasileiro de Administrar. Atlas, São Paulo, SP, Brasil.

Bioni, B.R. 2021. Proteção de Dados Pessoais: a função e os limites do consentimento. 2. ed. Forense, Rio de Janeiro, RJ, Brasil.

Bortolotti, S.L.V.; Moreira Junior, F.J.; Bornia, A.C. 2012. Avaliação da satisfação: um estudo dos instrumentos de medida com escala Likert. Produção Online, Florianópolis, 12(2): 440-462.

Caldas, M.P.; Wood Jr., T. 1999. Transformação e Realidade Organizacional: uma perspectiva brasileira. Atlas, São Paulo, SP, Brasil.

Cohen, J. 1988. Statistical Power Analysis for the Behavioral Sciences. 2. ed. Lawrence Erlbaum, Hillsdale, NJ, EUA.

Davenport, T.H.; Harris, J.G. 2007. Competindo em Analytics: a nova ciência de vencer. Elsevier, Rio de Janeiro, RJ, Brasil.

Diniz, E.H.; Barbosa, A.F.; Junqueira, A.R.B.; Prado, O. 2009. O governo eletrônico no Brasil: perspectiva histórica a partir de um modelo estruturado de análise. Revista de Administração Pública, Rio de Janeiro, 43(1): 23-48.

Field, A. 2009. Descobrindo a Estatística Usando o SPSS. Artmed, Porto Alegre, RS, Brasil.

Fleury, A.; Fleury, M.T.L. 2001. Estratégias Empresariais e Formação de Competências: um quebra-cabeça caleidoscópico da indústria brasileira. Atlas, São Paulo, SP, Brasil.

Hair, J.F.; Black, W.C.; Babin, B.J.; Anderson, R.E.; Tatham, R.L. 2009. Análise Multivariada de Dados. 6. ed. Bookman, Porto Alegre, RS, Brasil.

Hirano, H.I. 2024. O processo de tomada de decisões organizacionais com o uso do Data Analytics. Tese de Doutorado. Universidade de São Paulo, São Paulo, SP, Brasil. Disponível em: <https://www.teses.usp.br/teses/disponiveis/12/12139/tde-14012025-124449/>. Acesso em: 15 dez. 2025.

Hourneaux Junior, F.; Rueda Elias Spers, V.; Correa, H.L. 2005. Avaliação de desempenho organizacional: estudo de casos de empresas do setor químico. Revista de Administração da USP (RAUSP), São Paulo, 40(4): 364-378.

Kaplan, R.S.; Norton, D.P. 1997. A Estratégia em Ação: Balanced Scorecard. Campus, Rio de Janeiro, RJ, Brasil.

Kaufman, L.; Rousseeuw, P.J. 1990. Finding Groups in Data: An Introduction to Cluster Analysis. John Wiley & Sons, New York, NY, EUA.

Laurindo, F.J.B.; Rotondaro, R.G. (org.). 2006. Gestão Integrada de Processos e da Tecnologia da Informação. Atlas, São Paulo, SP, Brasil.

Malhotra, N.K. 2019. Pesquisa de Marketing: uma orientação aplicada. 7. ed. Bookman, Porto Alegre, RS, Brasil.

Martins, R.A.; Salerno, M.S. 1998. Sistemas de medição de desempenho: uma revisão da literatura. Gestão & Produção, São Carlos, 5(2): 136–158.

Nonaka, I.; Takeuchi, H. 1995. The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation. Oxford University Press, New York, NY, EUA.

Pereira, F.M. 2023. Cultura orientada a dados e Inteligência de negócios: avaliação da maturidade organizacional. Tese de Doutorado em Administração. Universidade Federal do Rio Grande do Norte, Natal, RN, Brasil.

Podsakoff, P.M.; MacKenzie, S.B.; Lee, J.-Y.; Podsakoff, N.P. 2003. Common method biases in behavioral research: a critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5): 879–903.

Richardson, H.A.; Simmering, M.J.; Sturman, M.C. 2009. A tale of three perspectives: examining post hoc statistical techniques for detection and correction of common method variance. Organizational Research Methods, 12(4): 762–800.

Schein, E.H. 2010. Organizational Culture and Leadership. 4th ed. Jossey-Bass, San Francisco, CA, EUA.

Schwab, K. 2016. A Quarta Revolução Industrial. Edipro, São Paulo, SP, Brasil.

Sordi, J.O.; Meireles, M. 2011. Análise de informações qualitativas: procedimentos e suas aplicações em pesquisas usando a técnica Delphi. Revista de Administração Mackenzie (RAM), São Paulo, 12(4): 67–97.

Teece, D. J. 2007. Explicating dynamic capabilities: the nature and microfoundations of (sustainable) enterprise performance. Strategic Management Journal, 28(13), 1319-1350.

Article originating from the Final Course Project of the Specialization in Data Science and Analytics of the MBA USP/Esalq

To learn more about the course, click here and access the MBX Academy platform