Business model experimentation in micro and small enterprises: integrating Business Model Canvas and System Dynamics

Abstract

Paper aims  The static nature of the Business Model Canvas (BMC) limits strategic decision-making in dynamic environments, particularly for micro and small enterprises (MSEs). This study develops and validates a dynamic framework integrating the BMC and System Dynamics (SD) to support business model experimentation over time.

Originality  The research transforms the BMC into a dynamic simulation environment by incorporating feedback loops, delays, and non-linear relationships.

Research method  A mixed-method approach was adopted. A systematic literature review identified key business model variables and relationships. Based on the nine BMC building blocks, a System Dynamics model was developed, validated by experts, and applied to scenario analyses involving different business models, market growth conditions, and input price variations.

Main findings  The simulations reveal dynamic effects not captured by static representations. B2B models showed lower but more stable performance under adverse conditions, whereas B2C models achieved higher profitability under favorable scenarios but greater sensitivity to input price volatility.

Implications for theory and practice  The framework integrates business model design and systems thinking while providing entrepreneurs with a decision-support tool to test strategic alternatives under uncertainty. Future research should apply the model in real-world settings to evaluate its effects on entrepreneurial decision-making and firm survival.

Keywords:
Business Model Canvas; System Dynamics; Business model experimentation; Scenario analysis; Decision support

1. Introduction

The business model concept is widely used to describe how organizations create, deliver, and capture value (Osterwalder & Pigneur, 2011). Business models play a central role in the pursuit of competitive advantage, as they support strategic choices that differentiate firms from their competitors (Martikainen et al., 2014). Business model innovation, in turn, refers to changes that reshape markets or alter key competitive factors (Euchner & Ganguly, 2014).

Among the available frameworks, the Business Model Canvas (BMC) has gained prominence due to its structured and intuitive representation of essential business model elements, such as value creation, delivery, and capture (Wrigley et al., 2016). Despite its widespread adoption by researchers and practitioners, the BMC is frequently applied in a static manner, limiting the analysis of interdependencies, feedback effects, and external influences that evolve over time (Ching & Fauvel, 2013; Khodaei & Ortt, 2019). As a result, several studies argue that complementary approaches are required to incorporate dynamic and systemic perspectives into business model analysis.

In this context, experimentation has emerged as a relevant mechanism to support the design and evaluation of business models, particularly under conditions of uncertainty (Bland et al., 2020; Brunswicker et al., 2013). Experimentation enables organizations to test strategic assumptions, reduce risk, and guide investment decisions before full-scale implementation. This issue is especially critical for micro and small enterprises, which face high failure rates, particularly in their early years of operation (Serviço Brasileiro de Apoio às Micro e Pequenas Empresas, 2021). However, the static nature of traditional BMC applications limits the ability of entrepreneurs to dynamically experiment with strategic alternatives and understand how business model elements interact over time, potentially contributing to ineffective strategic decisions under uncertainty.

To address these limitations, prior studies have explored the integration of dynamic methods into business model frameworks. Romero et al. (2015), for example, proposed the use of System Dynamics to enable the simulation of business model elements and their interactions over time. By translating the BMC into a System Dynamics model, it becomes possible to experiment with alternative policies and scenarios, capturing feedback loops, delays, and nonlinear relationships (Cosenz, 2017; Poles, 2013).

System Dynamics is a well-established methodology for analyzing complex systems and supporting learning through simulation (Sterman, 2000). When applied to business models, it offers a promising approach for overcoming the limitations of static representations and enhancing strategic decision-making.

Based on these arguments, this research aims to propose and validate a System Dynamics-based simulation model that operationalizes the Business Model Canvas, thereby creating a dynamic environment for business model experimentation tailored to the context of micro and small enterprises. The study contributes to the literature by offering a dynamic framework that supports business model experimentation and provides practical insights for entrepreneurial learning and decision-making. The remainder of the paper is organized as follows: the next section presents the theoretical background, followed by the research design, results, discussion, and conclusions.

2. Theoretical background

Traditional business model representations emphasize the internal coherence among key elements and typically serve as static blueprints of organizational activities (Demil & Lecocq, 2010). Although useful for structuring strategic discussions, such representations often fail to capture dynamic interactions, feedback mechanisms, and temporal effects inherent to business models. This limitation is particularly evident in the conventional use of the Business Model Canvas (BMC), which is commonly constructed through static visual artifacts, such as sticky notes.

To overcome this limitation, Romero et al. (2015) proposed integrating System Dynamics into the BMC through the Enriched Canvas Meta-Model. This framework extends the traditional BMC by incorporating dynamic elements that represent flows and transformation processes among business model components. By translating the BMC into a System Dynamics model, it becomes possible to simulate interactions among elements and explore the dynamic consequences of strategic decisions over time (Cosenz, 2017; Poles, 2013).

Building on this perspective, Cosenz (2017) introduced the Dynamic Business Model Canvas (DBMC), which combines the BMC with principles of Dynamic Performance Management. The DBMC emphasizes causal relationships among business model elements and integrates performance indicators to support longitudinal analysis. While maintaining the synthetic structure of the BMC, this approach enhances the systemic understanding of business performance dynamics.

Further advancing this line of research, Cosenz & Noto (2018) proposed the Dynamic Business Model (DBM), a simulation-based framework grounded in System Dynamics. The DBM supports the exploration of alternative strategies and external scenarios by modeling the long-term behavior of key performance variables. Although empirical applications demonstrate the potential of the DBM, prior studies report challenges related to model complexity, variable selection, and the need for facilitation during model development.

Recent research has extended dynamic business modeling approaches to contexts such as sustainability-oriented business models (Cosenz et al., 2020), digital platforms (Madanaguli et al., 2023), and decision support for micro and small enterprises (Bhatia & Diaz-Elsayed, 2023; Saka et al., 2022). Additionally, System Dynamics has been combined with multicriteria and fuzzy decision-making methods to support complex managerial decisions across different sectors (Orji & Liu, 2020; Santos et al., 2002). These studies illustrate the versatility of System Dynamics as a decision support tool, while also highlighting limitations related to usability, data assumptions, and model interpretation (Brugler et al., 2024).

Overall, the literature demonstrates the relevance of integrating dynamic mechanisms into business model frameworks to support experimentation and decision-making. However, existing approaches often lack clear guidance on model completeness and variable selection, particularly for entrepreneurs and small firms. This gap motivates the development of a structured dynamic experimentation model based on the Business Model Canvas, which is presented in the subsequent sections.

3. Research design

For this research, a quantitative modeling approach was selected. Models support understanding the business environment, identifying problems, formulating strategies, and supporting decision-making (Bertrand & Fransoo, 2002; Sinuany-Stern, 2023). Quantitative models use control and performance variables to evaluate decision quality through the causal and quantitative relationships among them. The model was constructed following the steps proposed by Bertrand and Fransoo (2002) were followed. The main stages of the process were defined as follows: problem definition discussed in the previous sections; model construction; model solution; model validation; and solution implementation.

3.1. Model construction

A systematic literature review (SLR) was conducted to identify the essential variables used to represent a BMC. The search was performed in the Scopus database using the term “Business Model Canvas”. The initial search returned 562 records. Titles and abstracts were screened according to the inclusion and exclusion criteria summarized in Table 1. During screening 351 records were excluded, and 211 studies were retained for full-text retrieval; 23 full texts could not be located. After full-text reading and application of the eligibility criteria, 93 documents constituted the final corpus used to extract and define the key variables for the model. The complete literature review protocol, including the search string, screening procedures, eligibility assessment, corpus selection flow is presented in Appendix A. The final corpus of articles included in the review is presented in Appendix B. These materials were included to enhance the transparency, traceability, and repeatability of the literature review process.

Table 1
Criteria of inclusion and exclusion.

Inclusion (I) or Exclusion (E)	Critery	Description
I-01	Feasibility Analysis	The study seeks to define, through the BMC, the feasibility of a business.
I-02	Business Model Assessment/Analysis	The study evaluates or analyzes a business model.
I-03	Experimentation or Simulation	The study aims to simulate or prototype a business model.
I-04	Performance	The study quantifies the performance of a business model.
I-05	Systems Thinking/System Dynamics	The study addresses the relationships among the elements of a business model.
I-06	Business Model Comparison	The study compares business models applied by companies in the same sector, area, or field.
E-01	Framework Comparison	The study compares different business modeling frameworks.
E-02	BMC Used for Description	The BMC is used only to describe a business or to propose a new model.
E-03	BMC Adaptation	The study proposes a new framework or methodology to solve a problem, regardless of how strongly the BMC influenced its creation.
E-04	Teaching	The study applies to the BMC to evaluate or develop teaching practices.
E-05	Sustainability/Circular Business Models	The study focuses on the development of sustainable or circular business models.
E-06	Innovative Business Models	The study explores ways to foster innovation in business models.
E-07	Other	The study is unrelated to the topic.
E-08	Non-BMC Related	The title and abstract do not mention the BMC, or author information is unavailable.

The selected studies were analyzed using content analysis. The initial screening was conducted by the lead researcher and subsequently validated by the research supervisor. The coding process combined a priori categorical codes and open (emergent) codes. Categorical codes were established before the analysis and represented predefined analytical dimensions, such as research area, problem class, evaluation type, software used, and metrics. Open codes emerged during the analysis of the selected studies and captured specific characteristics, methods, variables, and research problems identified in the literature.

Figure 1 presents the coding structure adopted in this study, including the main coding groups and representative examples of categorical and open codes. To enhance transparency, traceability, and reproducibility, the complete coding framework, including all codes, definitions, and the coding matrix used during the analysis, is available in the Data Support repository (Zenodo). Appendix C presents the final corpus of articles included in the systematic literature review.

Figure 1
Main coding groups and representative examples of categorical and open codes. Source: Authors themselves.

As an example of the coding procedure, the study by Martikainen et al. (2014) investigated how logistics service providers could contribute to the development of the local food sector in Southeast Finland. This research problem was initially recorded as the open code Pp08 (“Finding a way, through logistics service providers, to develop the local food sector in Southeast Finland”). Subsequently, this open code was associated with three predefined problem categories: Cp02 (Business Model Analysis), Cp03 (Supply Chain Improvement), and Cp04 (Generic Business Model Proposal). This procedure illustrates how qualitative evidence extracted from the literature was systematically transformed into structured analytical categories, enabling the identification of variables and relationships used in the model development process.

The identified codes were subsequently synthesized into variables associated with business model representation, evaluation, and experimentation. These variables were grouped according to four main business areas: Infrastructure, Value Proposition, Customers, and Financial Viability, which correspond to the nine elements of the Business Model Canvas (BMC). Figure 2 summarizes the resulting variables and their allocation across the BMC structure. The funnels represent the literature inputs, while the lower section illustrates how the identified variables were categorized and linked to the BMC building blocks.

Figure 2
Variables Identified in the Literature. Source: Authors themselves.

Figure 2 illustrates the synthesis process used to derive the final set of variables incorporated into the model. The synthesis process was not limited to grouping variables with similar meanings. Variables were critically evaluated regarding their ability to represent generic business models and support the development of a System Dynamics simulation model. Variables considered overly specific to particular industries, case studies, or business contexts were excluded, while conceptually overlapping variables were consolidated.

This procedure was performed separately for each of the four business areas associated with the Business Model Canvas. For example, in the Value Proposition area, ten variables were initially identified in the literature. However, several of these variables were excluded because they were specific to particular business contexts or represented concepts already captured by other variables. Given the limited number of generic variables available in the literature, additional variables were incorporated to better represent the Value Proposition dimension, including Degree of Innovation on the Value Proposition, Product-Market fit, Fit to the Problem Solution, and Adequacy of the Business Model. This process resulted in a final set of eight key variables: Quality of the Product, Price, Degree of Innovation on the Value Proposition, Degree of Customization of the Value Proposition, Product-Market fit, Fit to the Problem Solution, Adequacy of the Business Model, and Level of Innovation, capable of representing the Value Proposition area in a comprehensive and generic manner.

The overall model development process followed a structured sequence involving variable identification through a systematic literature review, conceptual modeling, and System Dynamics simulation, similarly to approaches adopted in previous studies (Martins et al., 2020).

3.2. Construction of the causal loop diagram

A preliminary causal loop diagram was developed from the key variables defined in the previous section and then validated through expert interviews and agreement evaluation. The essential variables were connected through systemic transcription, as proposed by Kim & Andersen (2012). Prior experience in designing business models for micro and small enterprises using the BMC was required for expert selection. The interviewees’ profiles are presented in Table 2.

Table 2
Profile of interviewees.

Position	Time in Position	Education	Age
Technical Manager	2 years and 6 months	Bachelor’s in Agronomy	34
Project Manager and Co-Founder	2 years and 10 months	Bachelor’s in Portuguese Language and currently pursuing Digital Design	31
Senior Business Management Consultant	2 years	Bachelor’s in production engineering	28

Following the identification and synthesis of the key variables, a preliminary causal loop diagram was developed based on the systemic transcription approach proposed by Kim & Andersen (2012). The identified variables served as the foundation for establishing cause-and-effect relationships using Systems Thinking principles and the researchers’ interpretation of the phenomenon under investigation. Rather than directly extracting every causal relationship from individual literature statements, the literature-derived variables were synthesized and integrated into a preliminary causal structure representing the main dynamics of business models in micro and small enterprises.

The resulting preliminary causal loop diagram was subsequently triangulated and validated through expert interviews. This validation process contributed to the refinement of causal relationships and the inclusion of additional variables considered relevant by the specialists. Inter-rater agreement on the proposed structure was measured using Fleiss’ Kappa, indicating fair agreement according to the classification proposed by Landis & Koch (1977). Although the Fleiss’ Kappa value indicates fair agreement, all experts assigned scores of 4 or 5 to the evaluated dimensions, demonstrating a consistently positive assessment of the proposed structure. Further details regarding the expert validation process, including the evaluation scale, questionnaire, individual ratings, and Fleiss’ Kappa calculations, are presented in Appendix C. In addition to the quantitative assessment, the specialists provided qualitative feedback regarding missing variables, omitted relationships, and potential improvements to the systemic structure.

To assess the adequacy of the proposed structure, the specialists answered three objective questions regarding the representativeness of the variables, the coverage of the nine Business Model Canvas building blocks, and the validity of the proposed causal relationships. The questions and responses are presented in Table 3.

Table 3
Interview questions.

Question	Specialist 1	Specialist 2	Specialist 3
(1) Do the variables displayed in the diagram represent the main variables of a business?	5	4	4
(2) Do the variables of the causal loop diagram represent the 9 elements of a CANVAS business model?	5	5	4
(3) Do the casual relationships present in the causal loop diagram represent the existing relationships among the variables of a business?	5	5	4

Together, the experts provided complementary perspectives covering entrepreneurial practice, business management, and consulting activities, which were considered sufficient to evaluate the financial, operational, and customer-related dimensions of the proposed framework. As shown in Table 3, the specialists expressed a high level of agreement regarding the adequacy of the proposed structure. The evaluated dimensions obtained scores ranging from 4 to 5, indicating that the identified variables represent the main elements of a business, adequately cover the nine Business Model Canvas building blocks, and capture the principal causal relationships expected in a business model.

A complementary open-ended question was also posed to the specialists, asking whether any variable or causal relationship should be excluded from the diagram. All interviewees agreed that no variable should be removed, as doing so could reduce the accuracy of the business model representation and compromise the credibility of the resulting analyses. Nevertheless, the specialists suggested improvements to enhance the representativeness of the structure.

Based on this feedback, the preliminary causal loop diagram was refined through the inclusion of additional variables and causal relationships. The main modifications involved the incorporation of variables related to employee turnover, absenteeism, stakeholder satisfaction, and supplier reliability, as well as their respective interactions with the existing structure. These refinements improved the representativeness of the final systemic structure.

The final causal loop diagram resulting from the literature synthesis, systemic transcription process, and expert validation is presented in Figure 3.

Figure 3
Final Causal Loop Diagram. Source: Authors themselves.

3.3. Scenario analysis

The scenario analysis in this research followed the ten-step framework proposed by Schoemaker (1991), which emphasizes the identification of key uncertainties, assessment of internal consistency, and development of plausible and distinct futures. The approach was adapted to the context of micro and small enterprises to explore how different strategic configurations and external conditions influence business model performance.

The time horizon for the analysis was set at five years, consistent with the typical planning period adopted by small enterprises. The main actors considered in the model included the entrepreneur, customers (retailers and end consumers), and governmental agents, representing relevant interactions for business development.

The identification of critical uncertainties followed the scenario-building framework proposed by Schoemaker (1991). Initially, the entrepreneur and the researchers identified the key performance variables, relevant stakeholders, and the main trends and predetermined elements capable of affecting the business environment. Based on this assessment, six potential sources of uncertainty were identified: petroleum-derived input prices, rainfall patterns, sugar consumption, market growth, national economic performance, and the business model configuration. These factors were evaluated according to their potential influence on the key performance variables of the model. As a result, three critical uncertainties were selected for scenario construction: (i) the type of business model adopted (B2B vs. B2C), (ii) market growth rate, and (iii) variation in petroleum-derived input prices.

Each uncertainty was analyzed through distinct states (e.g., growth or decline, stable or volatile prices) to generate combinations of possible futures. This structure enabled the simulation of eight consistent and plausible scenarios to test the robustness of business model alternatives under different external conditions. This systematic scenario-building process supported the dynamic simulation model, enabling an assessment of how strategic decisions interact with environmental variables and influence long-term outcomes.

3.4. Construction of the System Dynamics model

The System Dynamics model was built from the final causal loop diagram using Stella Architecture software. The validated causal relationships were translated into stock-and-flow structures, where stocks represent the accumulation of resources, capabilities, customers, and financial results over time; flows represent the rates of change of these accumulations; and auxiliary variables capture intermediate relationships, decision rules, and external influences. To facilitate model development and interpretation, the structure was organized into four interconnected modules corresponding to the main business areas identified in the literature: Financial Viability, Structure, Customers, and Value Proposition. The resulting model comprises four modules, presented in Figures 4 to 7. Figure 4 is represented as the module of the financial viability area of a business. Elements from other areas also influence financial viability. For example, the element of installed capacity is influenced by an element from the customers' area, demand, and another from the structure area, installed capacity. In this context, sales revenue is defined based on the price and quantity of products sold, which consists of the lesser value between installed capacity and demand. Since price is a complex variable influenced by many uncertainties, the model requires the user to input the market price. This increases the model’s reliability and allows simulation of different market prices.

Figure 4
System Dynamics Model – Financial Viability Area Module. Source: Authors themselves.

Figure 7
System Dynamics Model – Value Proposition Area Module. Source: Authors themselves.

Fixed and variable costs are also deemed essential for measuring profit. The cost of raw materials, shipping costs, and other variable costs must also be reported by the model user, and their total is summed up to account for the variable expenses of the period. In the model, fixed expenses include employees’ expenses, structure maintenance costs, and process efficiency. Employee expenses are calculated as the sum of two factors: (i) the product of the initial number of employees and their average cost (provided by the user), and (ii) the product of the additional employees for marginal production and the capacity increase. Subsequently, the user will be able to establish the effect of the valuation policy on employee expenses through a graphical input. Structure maintenance cost is represented as a percentage of the investment, allowing the entrepreneur to select the percentage that best fits their reality. It is important to highlight that, within the model, investment in structure consists of the sum of investments in capacity increase, service structure, and marketing structure. The user must also define the effect of process efficiency on fixed expenses through graphical input.

The model establishes the investment value for the subsequent period through the profit of the current period, the profit distribution percentage, and investor contributions. Thus, the profit distribution percentage and investor contributions need to have their rules established by the system user. The selected solution was to define a profit distribution percentage and create a graphical input that establishes the relationship between investor satisfaction and their contributions.

To assess the financial viability of the business, the model computes three key indicators: ROI, Payback, and NPV. The NPV calculation considers (i) the Initial Investment, provided by the user as a one-time cash outflow, (ii) the Profit of the Period, which is generated during the simulation and accumulated in the stock variable Accumulated Profit as cash inflows, and (iii) the Hurdle Rate, defined as 0.5% per month (≈ 6.17% per year in compound terms). Accordingly, the NPV is expressed as:

$NPV=-InitialInvestment+\frac{AccumulatedProfit}{{\left(1+HurdleRate\right)}^{TIME}}$ (1)

The profit for each period is also essential for evaluating investor satisfaction and is used, together with the initial investment and the hurdle rate, to calculate ROI and Payback.

It is important to emphasize that some adjustments were necessary for the construction of the module presented in Figure 3. For example, for each type of investment, a suggested value by an expert was established to indicate whether the investment made is well above the competition's investment; above the competition's investment; at the level of the competition's investment; below the competition's investment; or well below the competition's investment. In this way, quantitative variables can be converted into qualitative ones.

Figure 5 presents the module of the structure area. In this module, investment in capacity increases the installed capacity of the business, which, in turn, increases capacity utilization, provided that the production volume remains constant. Capacity utilization is one of the five factors related to planning complexity. These factors also include supply reliability, the degree of customization of the value proposition, turnover, and absenteeism, which collectively define planning complexity. Investment in skills and competencies in processes leads to an increase in process efficiency, which, in turn, is also influenced by planning complexity and affects available capacity.

Figure 5
System Dynamics Model – Structure Area Module. Source: Authors themselves.

The addition of variables, such as unit cost of capacity, construction time (to be determined by the model user), and investment in processes, was necessary for the construction of this module of the System Dynamics model. Capacity utilization is one of the five variables related to planning complexity. This relationship must be established by the entrepreneur in two stages. Initially, a conversion to a qualitative variable, named Likert Capacity Utilization, is necessary, which must be carried out through a graphical input. Subsequently, the influence of capacity utilization on planning complexity must be established through a percentage.

Figure 6 presents the module developed for the customer area. As can be observed, skills and competencies in customer service are developed through investment in this area. It is noted that the quality of service is defined by the skills and competencies of the team, service structure, turnover, and absenteeism. A more lasting relationship with the customer can be developed from quality service, resulting in greater referrals and more loyal customers. These are defined as individuals who have a long-term relationship with the company and frequently use its services. Skills and competencies in customer service are also related to the salespeople’s efficiency. The demand variable is composed directly or indirectly of all variables in the customer area.

Figure 6
System Dynamics Model – Customers Area Module. Source: Authors themselves.

Financial investment variables, such as investment in skills and competencies in customer service and investment in consumer acquisition, require additional variables for unit conversion. In this module, we added variables for training investment in customer service and the cost of acquiring a consumer. These enable the conversion of investments into the qualitative variable of skills and competencies in customer service and the quantitative variable of the number of consumers acquired, respectively. The entrepreneur can establish the value of the conversion variables according to their business.

It is important to note that these additional variables are not conceptual constructs created by the authors, but auxiliary conversion variables introduced to ensure dimensional consistency within the System Dynamics model. Their function is to translate financial investments into measurable effects on stock and flow variables. This type of auxiliary element is a common modeling practice in System Dynamics and does not alter the theoretical foundation derived from the literature. The specific values of these conversion variables can be established by the entrepreneur according to the characteristics of their business.

To calculate the quality of service, it was also necessary to introduce some new variables into the model, namely: the weight of customer service structure on service quality; the weight of absenteeism and turnover on service quality; and the weight of skills and competencies in customer service on service quality. The model user can establish the composition that best fits their reality through an interactive graph, which limits the sum of the weights to a total of 100%.

Figure 7 presents the module developed for the value proposition area. It can be observed that investments in skills and competencies in processes and in skills and competencies in innovation enable the development of innovation in the areas of production, business model, product, and marketing. Innovation in any of these areas, along with the quality of the offered product and the degree of customization of the value proposition, allows the model to qualitatively indicate the level of fit to the problem solution. Thus, the proposed solution can be classified as: much more adequate than the competition; more adequate than the competition; adequate at the level of the competition; less adequate than the competition; or much less adequate than the competition.

For the conversion of investments in skills and competencies in innovation into the qualitative variable of development programs, the addition of the variable 'investment in innovation' (suggested by experts) was necessary. In this context, the organization's development program for skills and competencies can be classified as well above the competition, above the competition, at the level of the competition, below the competition, and well below the competition. The variable for development programs in skills and competencies in processes was developed similarly, as described in the presentation of the structure area.

It is important to clarify some points regarding the quality of raw materials. This variable relates to product quality and variable costs. Some interviewees pointed out that the relationship between the quality of raw materials and the unit cost is not always linear—i.e., higher quality does not always mean higher costs. However, this proposition is true for most scenarios. Since the model aims to be generic, allowing application in different sectors, it was decided to leave this relationship in the final causal loop diagram. Through the construction of the System Dynamics model, it was possible to exclude this relationship while still accommodating all possible scenarios. Therefore, the model allows the user to input the unit cost of the raw materials they intend to use and to define whether the quality of these raw materials is much higher, higher, at the same level, worse, or much worse than that offered by the competition. Subsequently, the model will use the cost of raw materials to establish the profit for the period and, consequently, the financial indicators, while their quality will be used to establish product quality.

The model was iteratively verified during its implementation in Stella Architecture to ensure the consistency of equations, parameter values, and interactions among modules. However, formal validation procedures such as sensitivity analysis and extreme-condition testing were not performed and constitute opportunities for future research. The complete Stella model, including all equations, lookup functions, parameter values, and scenario configurations, is provided as Supporting Information to facilitate transparency, replication, and future model extensions.

4. Results

4.1. Scenario analysis and results

This section presents the scenario analysis developed to assess the dynamic behavior of the business model under different external conditions and strategic configurations. The scenario analysis followed the ten-step framework proposed by Schoemaker (1991), adapted to the context of a small dairy-based venture to ensure systematic identification of uncertainties, consistency checks, and plausibility testing. The time horizon was set at five years, in line with the planning period typically adopted by small enterprises, and the analysis included financial and non-financial variables: Net Present Value (NPV), Return on Investment (ROI), payback, process efficiency, customer satisfaction, and fit to the problem solution.

Based on the literature review and expert inputs, three critical uncertainties were identified as drivers for the construction of the scenarios: (i) the type of business model adopted (B2B or B2C), (ii) the variation in market growth, and (iii) the fluctuation of petroleum-derived input prices, which directly affect production and logistics costs. The combination of these uncertainties generated eight distinct and plausible scenarios, each representing a potential future context for the enterprise (Figure 8).

Figure 8
Scenarios. Source: Authors themselves.

Consequently, eight scenarios were constructed, as follows. All values of the exogenous variables used for the scenarios can be consulted in the Supporting information.

1. Scenario 1: Gas inflation of 0.6% per month; gasoline inflation of 0.5% per month; market contraction of 0.3% per year; Business Model 1;

2. Scenario 2: Gas inflation of 0.6% per month; gasoline inflation of 0.5% per month; market growth of 8.3% per year; Business Model 1;

3. Scenario 3: Gas deflation of 0.21% per month; gasoline deflation of 0.26% per month; market contraction of 0.3% per year; Business Model 1;

4. Scenario 4: Gas deflation of 0.21% per month; gasoline deflation of 0.26% per month; market growth of 8.3% per year; Business Model 1;

5. Scenario 5: Gas inflation of 0.6% per month; gasoline inflation of 0.5% per month; market contraction of 0.3% per year; Business Model 2;

6. Scenario 6: Gas inflation of 0.6% per month; gasoline inflation of 0.5% per month; market growth of 8.3% per year; Business Model 2;

7. Scenario 7: Gas deflation of 0.21% per month; gasoline deflation of 0.26% per month; market contraction of 0.3% per year; Business Model 2;

8. Scenario 8: Gas deflation of 0.21% per month; gasoline deflation of 0.26% per month; market growth of 8.3% per year; Business Model 2.

All scenarios were tested for internal consistency and plausibility using historical data for input prices and national consumption trends. No scenario was excluded. The dynamic simulation model was executed for each configuration, allowing the observation of how internal business structure and external forces interact over time. To ensure transparency and reproducibility, the complete set of exogenous variables, parameter values, lookup functions, and scenario configurations adopted in the simulations is available in the Data Support repository (Zenodo), allowing readers to inspect, replicate, and adapt the model.

The simulation results demonstrate that the business performance is highly sensitive to external conditions. Scenarios combining market growth with deflationary tendencies in petroleum-derived inputs presented the best financial outcomes. In particular, scenarios 2, 4, and 8 showed positive NPV and ROI, and shorter payback periods, indicating financial feasibility under favorable conditions. In contrast, scenarios characterized by market contraction and inflationary trends (e.g., scenarios 3 and 7) presented sharp declines in profitability and cash flow, highlighting the vulnerability of small businesses to adverse combinations of market and cost pressures.

The behavior of profit over the 60-month horizon is illustrated in Figure 9. The curves highlight how the combination of market growth and petroleum price variation drives the overall financial trajectory of the business. Scenarios 4 and 8, which combine market growth with deflationary trends in input prices, present sustained and significant profit increases after the second year, representing the most favorable conditions. Scenario 2 also shows a positive trajectory, although more moderate. Conversely, scenarios 1, 3, 5, 6, and 7 exhibit persistent losses, particularly under the simultaneous effect of market contraction and input inflation.

Figure 9
Profit Curve by Scenario. Source: Authors themselves.

Comparing the two business models, Business Model 1 (B2B – scenarios 1–4) displays lower but more stable profit trajectories, supported by recurring sales and reduced acquisition costs. Business Model 2 (B2C – scenarios 5–8), on the other hand, achieves higher profitability under favorable conditions (e.g., scenario 8) but also experiences greater volatility and deeper losses under adverse conditions (e.g., scenario 5). This reinforces the trade-off between stability and growth potential across strategic configurations.

Beyond financial indicators, the model captures relevant non-financial outcomes, reflecting the systemic feedback between internal processes and customer experience. Figure 10 shows the evolution of process efficiency across scenarios, where efficiency tends to improve under stable market conditions and deteriorate under cost pressures.

Figure 10
Process Efficiency Curves by Scenario. Source: Authors themselves.

Figure 11 presents the customer satisfaction curves, which are positively influenced by market growth and negatively affected by cost inflation and reduced service levels. Figure 11 illustrates the fit to the problem solution variable, which integrates the systemic effects of financial performance, process efficiency, and customer satisfaction.

Figure 11
Customer Satisfaction Level Curves by Scenario. Source: Authors themselves.

Taken together, these results emphasize how dynamic interactions between external variables and strategic choices shape overall business performance. The scenario analysis demonstrates that integrating System Dynamics and structured scenario building enables the exploration of feedback effects and learning about business model resilience under uncertainty. This combination allows the identification of which strategic configurations are robust across a range of possible futures, providing a richer understanding of business model behavior than static approaches.

The same ‘snowball’ effect related to the profitability of the business can be observed in the other non-financial variables under analysis. The reinforcing profitability dynamics observed in the simulations resemble the “Success to the Successful” systems archetype described by Senge (1990). In this structure, superior performance generates additional resources that can be reinvested to further improve capabilities and outcomes, creating a self-reinforcing growth cycle. The results suggest that business models capable of generating early financial surpluses may benefit from cumulative advantages over time, particularly through investments in customer acquisition, operational capabilities, and innovation activities.

Both the ‘Customer Satisfaction Level’ variable (Figure 12) and ‘Fit to the Problem Solution’ variable (Figure 13) converge after a certain point into two clusters: one formed by the financially viable scenarios and another by the unviable ones. This pattern suggests that the evolution of non-financial performance is strongly conditioned by the firm's ability to generate and reinvest resources over time. Such behavior is also consistent with the Dynamic Capabilities perspective by Teece et al. (1997), which argues that organizational performance emerges from the continuous development and reconfiguration of capabilities rather than from isolated strategic decisions. In this sense, customer satisfaction and problem-solution fit appear not only as outcomes of the business model but also as reinforcing mechanisms that shape its long-term viability.

Figure 12
Customer Satisfaction Level Curves by Scenario. Source: Authors themselves.

Figure 13
Fit to the Problem Solution Curves by Scenario. Source: Authors themselves.

4.2. Discussion of results

The research contributes to the understanding of the Business Model Canvas (BMC) and its experimentation. The proposed model's main contribution is the addition of a dynamic component to the experimentation of Canvas business models. In this regard, through System Dynamics, it was possible to establish and consider the interrelationships between the variables. Thus, this research adds a dynamic component to the tool proposed by Osterwalder & Pigneur (2011).

The scenario analysis revealed that external conditions, such as market growth and input price dynamics, significantly affected both financial and non-financial indicators. The discussion therefore integrates these effects, emphasizing that Business Model 1 tends to maintain stability in adverse environments, while Business Model 2 performs better in expanding markets but is more sensitive to input volatility.

The research shows that it is possible to enhance the completeness of the BMC by listing the variables that represent the four areas of a business, based on the coding conducted in the literature review. Additionally, validation of the System Dynamics model (constructed from these variables) was carried out through interviews with experts. In this context, experienced professionals were invited to analyze the variables of the causal loop diagram and suggest important elements that were missing from the model, such as stakeholder satisfaction, turnover, and absenteeism.

This research aligns with the work of Romero et al. (2015), Cosenz (2017), and Cosenz & Noto (2018) by using the BMC as the foundation for constructing a System Dynamics model. It advances by establishing the generic variables that users should consider based on a literature review and expert contributions. Thus, the proposed model prevents important variables from being overlooked and enhances the accuracy of experimentation.

These findings reinforce previous arguments by Cosenz & Noto (2018) regarding the importance of testing business model robustness under uncertainty. However, this study extends their work by explicitly integrating both financial and operational indicators within a scenario-based simulation framework, which allows the evaluation of trade-offs between profitability, efficiency, and customer satisfaction.

The use of generic variables also allows for the establishment of causal relationships without user involvement. This facilitates the model's use and reduces the need for knowledge in the field of System Dynamics from entrepreneurs. In this context, this research advances beyond Cosenz (2017), Cosenz & Noto (2018), and Cosenz et al. (2020) by proposing a generic, ready-to-use model instead of a structure that needs to be developed ad hoc, where a model with different variables is constructed for each application.

From a theoretical perspective, this contribution extends beyond improving model usability. By proposing a generic and transferable System Dynamics structure grounded in the Business Model Canvas, this study contributes to reducing the barriers associated with the application of Systems Thinking in entrepreneurship and business model experimentation. Previous studies have demonstrated the value of simulation for analyzing specific business models; however, their application typically requires the development of a new model for each context. In contrast, the framework proposed here provides a common structure that can be adapted and extended across different applications.

Furthermore, the adoption of a standardized modeling framework contributes to the cumulative development of knowledge in the field. Rather than developing entirely new models for each context, researchers can use a shared structure to compare business models, evaluate alternative strategies, and investigate how different parameters influence performance across industries and scenarios. In this sense, the proposed framework represents a step toward more transferable and comparable approaches to business model experimentation.

Another theoretical contribution was the use of the scenario construction and analysis technique proposed by Schoemaker (1991). Thus, scenarios could be constructed following a tested and approved methodology in the scientific community. The System Dynamics model was used to simulate the scenarios, allowing the evaluation of different business model proposals, as well as the performance of these proposals against varying behaviors of critical factors over time. These two dimensions had not been explored in the scenario analyses presented in the literature until now.

Thus, Cosenz (2017) assessed the behavior of the proposed business against divergent reinvestment compositions, while Cosenz & Noto (2018) composed scenarios based on the percentage invested in marketing. This research went further, showing that the proposed model allows for the evaluation of various business models targeted at different customer segments and with different cost compositions.

The developed model enables dynamic experimentation of a BMC. It aims to help entrepreneurs simulate their business behavior, considering the interrelationships between the variables that compose it. Thus, entrepreneurs can learn about their business's behavior in response to changes in financial and non-financial variables, such as ‘Process Efficiency’, ‘Customer Satisfaction Level’, and ‘Value Proposition Adequacy’.

Beyond supporting quantitative evaluation, the simulation contributes to managerial learning by enabling the visualization of how strategic policies interact with environmental dynamics. For instance, entrepreneurs can recognize that increasing marketing investments in B2C models is only effective under growth and stable cost conditions, whereas capacity investments are more critical for maintaining profitability in B2B settings.

The simulation results also highlight important managerial trade-offs that may not be evident through intuitive analysis alone. For example, the scenario analysis demonstrated that Business Model 2 achieved superior performance under favorable market growth conditions but exhibited greater sensitivity to increases in input costs. In contrast, Business Model 1 generated lower returns during expansion periods but maintained more stable performance under adverse conditions. These findings illustrate that business model selection should not be based exclusively on expected profitability but also on the entrepreneur’s risk tolerance and expectations regarding future market conditions. By making these trade-offs explicit, the simulation supports more informed strategic decision-making and reduces reliance on intuition-based judgments.

The model also enables entrepreneurs to verify the effects of selected policies through simulation after a certain period. It is common for policies developed to address a problem to show short-term improvements in indicators but a marked deterioration in the long term, due to the interrelationships between elements, resulting in a situation worse than the initial one (Sterman, 2000). Using the model, these policies can be tested before decision-making.

Another significant contribution to the entrepreneur is the ability to examine various business models over time. This allows for the exploration, experimentation, and enhancement of different ways to create value for the customer before the business is established. Furthermore, by constructing scenarios, entrepreneurs can assess the behavior of the business model(s) under various circumstances. In the application conducted in this research, it was possible to evaluate the behavior of two business models proposed by the entrepreneur, and after analyzing the model's results, the entrepreneur could verify that, in the case analyzed, there was no single best business model but rather different risk/return relationships. This process provides learning for the entrepreneur in decision-making.

Another significant aspect of support for entrepreneurs provided by the model developed in this research is the definition of the variables that make up a business model. Since the BMC is often created by entrepreneurs freely using sticky notes, there is a risk that important variables for representing the business may be omitted. In this context, the proposed model requires that important variables, defined based on a literature review and interviews with experts from all areas of the company, be considered.

Mapping the tools employed for constructing and evaluating a BMC is also valuable for the entrepreneur. In this context, the tools used for building the model provide robustness and speed to its construction. The tools used for evaluating the business model, such as SWOT analysis and interviews, enable the entrepreneur to develop new viable models, which can be tested through the model proposed in this research.

Micro and small enterprises (MSEs) play a crucial role in the Brazilian economy, representing the majority of active firms and a significant share of employment generation. According to Serviço Brasileiro de Apoio às Micro e Pequenas Empresas (2024), these enterprises account for approximately 27% of Brazil’s Gross Domestic Product (GDP) and 52% of formal employment in the country. However, their survival rate remains low due to limited access to analytical tools and structured decision-making support. The simulation-based model proposed in this study offers an accessible and adaptable decision-support tool for this segment, enabling entrepreneurs to explore strategic alternatives, evaluate business model robustness, and anticipate the consequences of decisions under uncertainty. The intended societal value of the proposed framework lies in its potential to support the sustainability and competitiveness of small enterprises through more informed decision-making. By helping entrepreneurs identify viable business models and evaluate strategic trade-offs before implementation, the model may contribute to stronger firms, which in turn can support economic stability, job creation, and local development.

5. Conclusions

This research successfully achieved its objective of developing a dynamic simulation model for Business Model Canvas experimentation. By integrating the Business Model Canvas, Systems Thinking, and System Dynamics, the study addresses one of the main limitations of traditional business model representations: their static nature. The proposed framework enables the analysis of how business model elements interact and evolve over time, allowing entrepreneurs and researchers to evaluate alternative business model configurations under different environmental conditions and strategic scenarios.

From a methodological perspective, the study contributes a replicable process for transforming a qualitative and static business modeling framework into a quantitative and dynamic simulation model. The proposed approach combines a systematic literature review, content analysis, expert validation, and structured scenario construction based on Schoemaker (1991), providing a transparent pathway that can be replicated and adapted in future studies.

From a theoretical perspective, the study advances the understanding of business model dynamics by identifying, formalizing, and integrating a generic set of variables and causal relationships associated with the Business Model Canvas. Rather than focusing solely on the individual elements of the framework, the proposed model highlights how these elements interact over time through reinforcing and balancing feedback structures, providing a more systemic representation of business model behavior.

From a practical perspective, the research provides a ready-to-use simulation framework that can support business model experimentation in micro and small enterprises. By allowing users to configure variables, policies, and scenarios, the model functions as a decision-support environment in which alternative business strategies can be evaluated before implementation. This capability enables entrepreneurs to explore trade-offs, anticipate risks, and reduce uncertainty when making strategic decisions.

The proposed framework can be applied by both researchers and practitioners. Researchers may use the detailed methodological procedure and the generic simulation structure as a basis for replication studies in different industries, countries, or organizational contexts, as well as for testing specific theories related to business model dynamics. Practitioners may use the model as a business “flight simulator”, exploring the potential effects of changes in market growth, input prices, customer behavior, or investment policies before committing resources in real-world settings.

Future research may further strengthen the model through additional validation procedures, such as sensitivity analysis and extreme-condition testing, as well as through applications in different sectors and business environments. To support transparency, reproducibility, and future developments, the complete System Dynamics model, including stock-and-flow structures, equations, lookup functions, parameter values, and scenario configurations, is publicly available in the Zenodo repository referenced in the Data Availability section. The dataset generated during the systematic literature review, together with the coding matrix and the materials used in the model construction process, is also available in the same repository, allowing researchers and practitioners to inspect, replicate, adapt, and extend the proposed framework.

Data availability

To support transparency, reproducibility, and future research, the complete System Dynamics model developed in this study, including stock-and-flow structures, equations, lookup functions, parameter values, and scenario configurations, is publicly available in the Zenodo repository. The repository also includes the dataset of variables extracted from the systematic literature review, the coding matrix, and the materials used in the model construction process. These resources allow readers to inspect, replicate, adapt, and extend the proposed framework for research and practical applications. Data Support repository: https://doi.org/10.5281/zenodo.20586767

Appendix A Systematic literature review protocol.

Figure A1
Flowchart of the data reduction process.

Appendix B Final corpus of studies included in the systematic literature review.

ID	Title	Authors / year	Source
p1	An approach to business model innovation and design for strategic sustainable development	França et al. (2017)	Journal of Cleaner Production
p2	Business innovation and government regulation for the promotion of electric vehicle use: lessons from Shenzhen, China	Li et al. (2016)	Journal of Cleaner Production
p3	Integrative re-use systems as innovative business models for devising sustainable product-service-systems	Gelbmann & Hammerl (2015)	Journal of Cleaner Production
p4	A review of telemedicine business models	Chen et al. (2013)	Telemedicine and e-Health
p5	Business model canvas perspective on big data applications	Muhtaroglu et al. (2013)	IEEE International Conference on Big Data
p6	Business model innovation in small- and medium-sized enterprises: Strategies for industry 4.0 providers and users	Müller (2018)	Journal of Manufacturing Technology Management
p7	Business models for model businesses: Lessons from renewable energy entrepreneurs in developing countries	Gabriel & Kirkwood (2016)	Energy Policy
p8	Developing a service offering for a logistical service provider-Case of local food supply chain	Martikainen et al. (2014)	International Journal of Production Economics
p9	Evolution of photovoltaic business models: Overcoming the main barriers of distributed energy deployment	Horváth & Szabó (2018)	Renewable and Sustainable Energy Reviews
p10	Business models for maximising the diffusion of technological innovations for climate-smart agriculture	Long et al. (2016)	International Food and Agribusiness Management Review
p11	Social farming in Catalonia: Rural local development, employment opportunities and empowerment for people at risk of social exclusion	Guirado et al. (2017)	Journal of Rural Studies
p12	Dynamic business modeling for sustainability: Exploring a system dynamics perspective to develop sustainable business models	Cosenz et al. (2020)	Business Strategy and the Environment
p13	The Business Model Evaluation Tool for Smart Cities: Application to SmartSantander use cases	Díaz-Díaz et al. (2017)	Energies
p14	Business models in urban farming: A comparative analysis of case studies from Spain, Italy and Germany	Pölling et al. (2017)	Moravian Geographical Reports
p15	Airline categorisation by applying the business model canvas and clustering algorithms	Urban et al. (2018)	Journal of Air Transport Management
p16	Methodology for exploiting potentials of remanufacturing by reducing complexity for original equipment manufacturers	Widera & Seliger (2015)	CIRP Annals - Manufacturing Technology
p17	What can we learn from business models in the European forest sector: Exploring the key elements of new business model designs	Kajanus et al. (2018)	Forest Policy and Economics
p18	Challenges for business change in district heating	Lygnerud (2018)	Energy, Sustainability and Society
p19	Proposal for a method for business model performance assessment: Toward an experimentation tool for business model innovation	Batocchio et al. (2017)	Journal of Technology Management & Innovation
p20	The main transition management issues and the effects of environmental accounting on financial performance–with focus on cement industry	Fogarassy et al. (2018)	Administratie si Management Public
p21	How to keep a living lab alive?	Mastelic et al. (2015)	Info
p22	Designing new business models: blue sky thinking and testing	Wrigley et al. (2016)	Journal of Business Strategy
p23	Examining the business case and models for sustainable multifunctional edible landscaping enterprises in the Phoenix metro area	Robinson et al. (2017)	Sustainability
p24	Simulating the business model canvas using system dynamics	Romero et al. (2015)	10th Computing Colombian Conference
p25	Designing business models options for 'University of the Future'	Ibrahim & Dahlan (2016)	4th IEEE International Colloquium on Information Science and Technology (CIST)
p26	New business models for electric mobility	Campatelli et al. (2014)	IEEE International Electric Vehicle Conference
p27	Generating a business model canvas through elicitation of business goals and rules from process-level use cases	Salgado et al. (2014)	Perspectives in Business Informatics Research
p28	Sustainable business models–canvas for sustainability, evaluation method, and their application to additive manufacturing in aircraft maintenance	Cardeal et al. (2020)	Sustainability
p29	Business model as a base for building firms’ competitiveness	Koprivnjak & Peterka (2020)	Sustainability
p30	Business model for developing strategies of forest cooperatives. Evidence from an emerging business environment in Greece	Trigkas et al. (2019)	Journal of Sustainable Forestry
p31	Business models for climate services: An analysis	Larosa and Mysiak (2019)	Climate Services
p32	Analysis of lean manufacturing strategy using system dynamics modelling of a business model	Segura et al. (2020)	International Journal of Lean Six Sigma
p33	A decision support tool for business models analysis	Latora & Trapani (2018)	International Journal of the Analytic Hierarchy Process
p34	Business model exploration for software defined networks	Xu et al. (2017)	Software Business
p35	Managing digital transformation through hybrid business models	Endres et al. (2020)	Journal of Business Strategy
p36	Differences, constraints and key elements of providing local sharing economy services in different-sized cities: A Hungarian case	Czakó et al. (2019)	Resources
p37	Eating hamburgers slowly and sustainably: The fast food market in north-west Italy	Bonadonna et al. (2019)	Agriculture
p38	Street food: A tool for promoting tradition, territory, and tourism	Bonadonna et al. (2019)	Tourism Analysis
p39	Synergies between app-based car-related shared mobility services for the development of more profitable business models	Gilibert & Ribas (2019)	Journal of Industrial Engineering and Management
p40	Business development strategy of sago for food security	Makkarennu et al. (2018)	IOP Conf. Series: Earth and Environmental Science
p41	Analysis of entrepreneurship perception and business developmental strategy of silk in Wajo Regency, South Sulawesi, Indonesia	Kadir (2018)	International Journal of Law and Management
p42	Business model in marketplace industry using business model canvas approach: An e-commerce case study	Erlyana & Hartono (2017)	IOP Conference Series: Materials Science and Engineering
p43	Reflections on science gateways sustainability through the business model canvas: Case study of a neuroscience gateway	Gesing & Wilkins-Diehr (2015)	Concurrency and Computation: Practice and Experience
p44	Business model innovation at the bottom of the pyramid – A case of mobile money agents	Iheanachor et al. (2021)	Journal of Business Research
p45	Business models of FinTechs – Difference in similarity?	Laidroo et al. (2021)	Electronic Commerce Research and Applications
p46	Business Model Canvas as Diagnostic Tool of the Creation of Cultural Value: Conceptual and Exploratory approuch to the Case of Ferreira's de Castro Cultural Complex	Tavares & Ferreira (2020)	Iberian Conference on Information Systems and Technologies
p47	Information gap in value propositions of business models of language schools	Mazurek & Kulakowski (2020)	Faculty of Management
p48	Study on Business Model of Virtual Power Plant based on Osterwalder Business Model Canvas	Li et al. (2019)	IEEE 3rd International Electrical and Energy Conference
p49	Marketing Strategic and Competitive Positioning of Palm Sugar Business development	Makkarennu et al. (2019)	IOP Conf. Series: Earth and Environmental Science
p50	Business model analysis for the interaction between smart grid and mobile network operators	Salahaldin et al. (2019)	International Journal of Global Energy Issues
p51	Sustainability of Open Education through collaboration	Langen (2018)	The International Review of Research in Open and Distributed Learning
p52	Assessing bike sharing business model	Guyandi et al. (2017)	International Conference on Informatics and Computational Sciences
p53	Innovative approaches in forest management - The application of a business model to designing a small-scale forestry strategy	Richard et al. (2017)	Journal of Forest Science
p54	How to ensure the economic viability of an open data platform	Duval & Brasse (2014)	Procedia Computer Science
p55	Production of fuel from plastic waste: A feasible business	Fahim et al. (2021)	Polymers
p56	A Business-Model Approach on Strategic Flexibility of Firms in a Shifting Value Chain: The Case of Coffee Processors in Amadeo and Silang, Cavite, Philippines	Tan (2020)	Global Journal of Flexible Systems Management
p57	Research on the Influential Factors of the Success of 7-ELEVEn in Japan	Lyu (2021)	International Conference on New Energy Technology and Industrial Development
p58	A business model and cost analysis of automated platoon vehicles assisted by the Internet of things	Chen et al. (2020)	Proceedings of the Institution of Mechanical Engineers
p59	A Fuzzy Strategy Analysis Simulator for Exploring the Potential of Industry 4.0 in End of Life Aircraft Recycling	Keivanpour (2021)	Advances in Intelligent Systems and Computing
p60	Networked economic value creation in event tourism: An exploratory study of towns and smaller cities in the UK	Mark & Nana (2021)	Event Management
p61	Successful implementation of telemedicine depends on personal relations between company representatives and healthcare providers: A qualitative study of business models for Danish home telemonitoring	Korsgaard et al. (2021)	Health Services Management Research
p62	Exploring Second Life Applications for Electric Vehicle Batteries	Vu et al. (2020)	Advances in Transdisciplinary Engineering
p63	Analysis of the Business Model of Online Video Site iQiyi	Liang (2020)	Proceedings of the 2020 4th International Conference on E-Business and Internet
p64	An Online Catering Marketplace Business Plan: Kunyahku.id	Natashia et al. (2020)	International Conference on Engineering and Information Technology for Sustainable Industry
p65	Business model and innovation strategy of a Brazilian food company	Donadon & Santos (2020)	Estudios Gerenciales
p66	How social business innovates health care: two cases of social value creation leading to high-quality services	Bohnet-Joschko et al. (2019)	Journal of Public Health: From Theory to Practice
p67	Analysis of Business Development Strategies with Business Model Canvas Approach	Mustaniroh et al. (2020)	IOP Conf. Series: Earth and Environmental Science
p68	Formulation of strategies for developing local chocolate product Socolatte in Pidie Jaya	Afni et al. (2020)	IOP Conf. Series: Earth and Environmental Science
p69	Mapping of business potentials of Maha orange plantation using the Business Model Canvas and BCG matrix	Maha et al. (2020)	IOP Conf. Series: Earth and Environmental Science
p70	Innovative social business model development for organic rice commodity entrepreneur using business model canvas (BMC) (Case study: Gapoktan Simpatik, local farmers group entrepreneur in Cisayong, Tasikmalaya)	Purnomo et al. (2020)	IOP Conf. Series: Earth and Environmental Science
p71	An Integration of Business Model Canvas on Prioritizing Strategy: Case Study of Small Scale Nontimber Forest Product (NTFP) Enterprises in Indonesia	Makkarennu et al. (2021)	Small-scale Forestry
p72	Size matters – an analysis of business models and the financial performance of finnish wood-harvesting companies	Jylhä et al. (2020)	Silva Fennica
p73	Economic and financial viability evaluation model for the implementation of waste recycling plants construction in Brazilian municipalities	Gularte et al. (2020)	Engenharia Sanitaria e Ambiental
p74	Business model of learning platforms in sharing economy	Cornejo-Velazquez et al. (2020)	The Electronic Journal of e-Learning
p75	The Role of Information Technology in Fintech Innovation: Insights from the New York City Ecosystem	Mamonov (2020)	International Federation for Information Processing
p76	Exploring business models of nonprofit organizations	Perić et al. (2020)	Management: Journal of Contemporary Management Issues
p77	Business model canvas as an analytical tool for the evaluation of companies: Case study for the audiovisual industry in Bogota, Colombia	Ruiz-Ramirez et al. (2019)	South African Journal of Industrial Engineering
p78	Internet of Things Business Models: The RAWFIE Case	Papadopoulou et al. (2019)	International Federation for Information Processing
p79	PetGrab-linker services – A conceptual business model	Dahlan et al. (2018)	International Conference on Information and Communication Technology for the Muslim World (ICT4M 2018)
p80	Mobile application business plan to assist travel planning	Sanjaya et al. (2017)	International Conference on Information Management and Technology (ICIMTech)
p81	Comparative study of sharing economy business models in accommodation sector	Gatautis et al. (2018)	31st Bled eConference: Digital Transformation
p82	Client orientation of central power generation companies	Fedosova & Volkova (2018)	International Journal of Energy Sector Management
p83	Business model for e-funding in creative industries	Gui et al. (2017)	Proceedings of the 2017 International Conference on Information Technology
p84	Making a business case for LID treatment methods	Harne (2013)	Ports 2013: Success through Diversification
p85	E-Commerce in Online Business	Yunanto & Paizal (2019)	IOP Conf. Series: Materials Science and Engineering
p86	E-Commerce in Online Business	Vanhala & Saarikallio (2015)	International Journal of Computer Information Systems and Industrial Management Applications
p87	A study of value using the Business Model Canvas [Um estudo sobre valor utilizando o Business Model Canvas]	Vicelli & Tolfo (2017)	Revista Espacios
p88	Peculiarities of IoT-based business model transformations in SMEs	Varaniūtė et al. (2018)	International Conference on Electronic Busines
p89	Business model for occupational activation of elderly people via the service E-marketplace platform - Research study of Polish market	Kutera et al. (2017)	Multi Conference on Computer Science and Information Systems
p90	Development of business model in SMEs of ship component to improve competitiveness	Khoryanton et al. (2020)	Quality
p91	Comparative study of e-commerce ventures: Copycat enablers in business models	Haertel et al. (2020)	Proceedings of the 2nd International Conference on Finance, Economics, Management and IT Business
p92	Business development strategy with business model canvas approach at Pakdhe Mie chicken shop-Cimanggis, Depok	Marfuah et al. (2019)	International Journal of Scientific & Technology Research
p93	Trenggalek typical food diversification strategies for increasing competitiveness in the SDGS era by using a business model canvas	Sawitri & Suswati (2019)	International Journal of Innovation, Creativity and Change

Appendix C Expert validation materials.

Table C1
Expert validation scale.

Score	Interpretation
1	Strongly disagree
2	Disagree
3	Neither agree nor disagree
4	Agree
5	Strongly agree

Table C2
Expert validation questionnaire.

Question	Purpose
Q1. Do the variables included in the diagram represent the main variables of a business?	Variable representativeness
Q2. Do the variables included in the Systemic Structure represent the nine Business Model Canvas building blocks?	BMC coverage
Q3. Do the causal relationships included in the Systemic Structure represent the relationships observed in a business context?	Structural validity

Table C3
Expert ratings.

Question	Expert 1	Expert 2	Expert 3
Q1	5	4	4
Q2	5	5	4
Q3	5	5	4

Table C4
Fleiss’ Kappa agreement assessment.

Measure	Value
Number of experts	3
Number of items evaluated	3
Number of response categories	5
Fleiss’ Kappa (κ)	0.306
Agreement level	Fair Agreement

• How to cite this article:
  Stefano, L. S., Feitosa, P. P. B., Lacerda, D. P., Morandi, M. I. W. M., & Piran, F. S. (2026). Business model experimentation in micro and small enterprises: integrating Business Model Canvas and System Dynamics. Production, 36, e20260032. https://doi.org/10.14488/1980-5411.20260032.
• Financial Support
  This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001. This research was also supported by the CNPq and FAPERGS, Brazilian research agencies, under grant numbers 404140/2023-8 (CNPq) and 23/2551-0001849-7 (FAPERGS). CAPES, CNPq, and FAPERGS had no role in the study design, data collection, data analysis, interpretation of the results, or writing of the manuscript.
• Ethical Statement
  The study involved expert consultations for model validation. No personal or sensitive data were collected, and all participants voluntarily agreed to participate.

References

• Bertrand, J. W. M., & Fransoo, J. C. (2002). Operations management research methodologies using quantitative modeling. International Journal of Operations & Production Management, 22(2), 241-264. https://doi.org/10.1108/01443570210414338
  » https://doi.org/10.1108/01443570210414338
• Bhatia, P., & Diaz-Elsayed, N. (2023). Facilitating decision-making for the adoption of smart manufacturing technologies by SMEs via fuzzy TOPSIS. International Journal of Production Economics, 257, 108762. https://doi.org/10.1016/j.ijpe.2022.108762
  » https://doi.org/10.1016/j.ijpe.2022.108762
• Bland, D. J., Osterwalder, A., Smith, A., & Papadakos, T. (2020). Testing business ideas. Retrieved in 8 March 2026, from https://www.strategyzer.com/pages/hubspot-form-test
  » https://www.strategyzer.com/pages/hubspot-form-test
• Brugler, S., Gardezi, M., Dadkhah, A., Rizzo, D. M., Zia, A., & Clay, S. A. (2024). Improving decision support systems with machine learning: identifying barriers to adoption. Agronomy Journal, 116(3), 1229-1236. https://doi.org/10.1002/agj2.21432
  » https://doi.org/10.1002/agj2.21432
• Brunswicker, S., Wrigley, C., & Bucolo, S. (2013). Business model experimentation: what is the role of design-led prototyping in developing novel business models? In M. Curley & P. Formica (Eds.), The experimental nature of new venture creation (Innovation, Technology, and Knowledge Management, pp. 139-151). Cham: Springer. https://doi.org/10.1007/978-3-319-00179-1_13
  » https://doi.org/10.1007/978-3-319-00179-1_13
• Ching, H. Y., & Fauvel, C. (2013). Criticisms, variations and experiences with Business Model Canvas. European Journal of Agriculture and Forestry Research, 1(2), 26-37.
• Cosenz, F. (2017). Supporting start-up business model design through system dynamics modelling. Management Decision, 55(1), 57-80. https://doi.org/10.1108/MD-06-2016-0395
  » https://doi.org/10.1108/MD-06-2016-0395
• Cosenz, F., & Noto, G. (2018). A dynamic business modelling approach to design and experiment new business venture strategies. Long Range Planning, 51(1), 127-140. https://doi.org/10.1016/j.lrp.2017.07.001
  » https://doi.org/10.1016/j.lrp.2017.07.001
• Cosenz, F., Rodrigues, V. P., & Rosati, F. (2020). Dynamic business modeling for sustainability: exploring a system dynamics perspective to develop sustainable business models. Business Strategy and the Environment, 29(2), 651-664. https://doi.org/10.1002/bse.2395
  » https://doi.org/10.1002/bse.2395
• Demil, B., & Lecocq, X. (2010). Business model evolution: in search of dynamic consistency. Long Range Planning, 43(2–3), 227-246. https://doi.org/10.1016/j.lrp.2010.02.004
  » https://doi.org/10.1016/j.lrp.2010.02.004
• Euchner, J., & Ganguly, A. (2014). Business model innovation in practice: a systematic approach to business model innovation can help capture value and reduce risks. Research Technology Management, 57(6), 33-39.
• Khodaei, H., & Ortt, R. (2019). Capturing dynamics in business model frameworks. Journal of Open Innovation, 5(1), 8. https://doi.org/10.3390/joitmc5010008
  » https://doi.org/10.3390/joitmc5010008
• Kim, H., & Andersen, D. F. (2012). Building confidence in causal maps generated from purposive text data: mapping transcripts of the Federal Reserve. System Dynamics Review, 28(4), 311-328. https://doi.org/10.1002/sdr.1480
  » https://doi.org/10.1002/sdr.1480
• Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data. Biometrics, 33(1), 159-174. https://doi.org/10.2307/2529310 PMid:843571.
  » https://doi.org/10.2307/2529310
• Madanaguli, A., Parida, V., Sjödin, D., & Oghazi, P. (2023). Literature review on industrial digital platforms: a business model perspective and suggestions for future research. Technological Forecasting and Social Change, 194, 122606. https://doi.org/10.1016/j.techfore.2023.122606
  » https://doi.org/10.1016/j.techfore.2023.122606
• Martikainen, A., Niemi, P., & Pekkanen, P. (2014). Developing a service offering for a logistical service provider-Case of local food supply chain. International Journal of Production Economics, 157(1), 318-326. https://doi.org/10.1016/j.ijpe.2013.05.026
  » https://doi.org/10.1016/j.ijpe.2013.05.026
• Martins, J. C., Lacerda, D. P., Wolf Motta Morandi, M. I., Goldmeyer, D., & Piran, F. S. (2020). Systems dynamic modeling to analyze the systemic viability of the combined use of regenerative converters and energy storage in a UPS manufacturer. Journal of Cleaner Production, 246, 118950. https://doi.org/10.1016/j.jclepro.2019.118950
  » https://doi.org/10.1016/j.jclepro.2019.118950
• Orji, I. J., & Liu, S. (2020). A dynamic perspective on the key drivers of innovation-led lean approaches to achieve sustainability in manufacturing supply chain. International Journal of Production Economics, 219, 480. https://doi.org/10.1016/j.ijpe.2018.12.002
  » https://doi.org/10.1016/j.ijpe.2018.12.002
• Osterwalder, A., & Pigneur, Y. (2011). Business Model Generation: inovação em modelos de negócios (Vol. 3). Rio de Janeiro: Alta Books.
• Poles, R. (2013). System Dynamics modelling of a production and inventory system for remanufacturing to evaluate system improvement strategies. International Journal of Production Economics, 144(1), 189-199. https://doi.org/10.1016/j.ijpe.2013.02.003
  » https://doi.org/10.1016/j.ijpe.2013.02.003
• Romero, M. C., Villalobos, J., & Sanchez, M. (2015). Simulating the business model canvas using system dynamics. In 2015 10th Colombian Computing Conference, 10CCC 2015 (pp. 527-534), Bogota, Colombia. New York: IEEE. https://doi.org/10.1109/ColumbianCC.2015.7333469.
• Saka, A. B., Chan, D. W. M., & Wuni, I. Y. (2022). Knowledge-based decision support for BIM adoption by small and medium-sized enterprises in developing economies. Automation in Construction, 141, 104407. https://doi.org/10.1016/j.autcon.2022.104407
  » https://doi.org/10.1016/j.autcon.2022.104407
• Santos, S. P., Belton, V., & Howick, S. (2002). Adding value to performance measurement by using system dynamics and multicriteria analysis. International Journal of Operations & Production Management, 22(11), 1246-1272. https://doi.org/10.1108/01443570210450284
  » https://doi.org/10.1108/01443570210450284
• Schoemaker, P. J. H. (1991). When and how to use scenario planning: a heuristic approach with illustration. Journal of Forecasting, 10(6), 549-564. https://doi.org/10.1002/for.3980100602
  » https://doi.org/10.1002/for.3980100602
• Senge, P. M. (1990). The fifth discipline: the art and practice of the learning organization. New York: Currency Doubleday.
• Serviço Brasileiro de Apoio às Micro e Pequenas Empresas – SEBRAE. (2021). A sobrevivência das empresas. Brasília.
• Serviço Brasileiro de Apoio às Micro e Pequenas Empresas – SEBRAE. (2024). Micro e pequenas empresas geram 27% do PIB do Brasil. Brasília. Retrieved in 8 March 2026, from https://sebrae.com.br/sites/PortalSebrae/ufs/mt/noticias/micro-e-pequenas-empresas-geram-27-do-pib-do-brasil,ad0fc70646467410VgnVCM2000003c74010aRCRD
  » https://sebrae.com.br/sites/PortalSebrae/ufs/mt/noticias/micro-e-pequenas-empresas-geram-27-do-pib-do-brasil,ad0fc70646467410VgnVCM2000003c74010aRCRD
• Sinuany-Stern, Z. (2023). Foundations of operations research: from linear programming to data envelopment analysis. European Journal of Operational Research, 306(3), 1069-1080. https://doi.org/10.1016/j.ejor.2022.10.046
  » https://doi.org/10.1016/j.ejor.2022.10.046
• Sterman, J. D. (2000). Business Dynamics Systems: thinking and modeling for a complex world Boston: Irwin/McGraw-Hill.
• Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabilities and strategic management. Strategic Management Journal, 18(7), 509-533. https://doi.org/10.1002/(SICI)1097-0266(199708)18:7<509::AID-SMJ882>3.0.CO;2-Z
  » https://doi.org/10.1002/(SICI)1097-0266(199708)18:7<509::AID-SMJ882>3.0.CO;2-Z
• Wrigley, C., Bucolo, S., & Straker, K. (2016). Designing new business models: blue sky thinking and testing. The Journal of Business Strategy, 37(5), 22-31. https://doi.org/10.1108/JBS-04-2015-0041
  » https://doi.org/10.1108/JBS-04-2015-0041

Edited by

• Editor(s)
  Adriana Leiras, Rodrigo Caiado

Publication Dates

• Publication in this collection
  10 Aug 2026
• Date of issue
  2026

History

• Received
  08 Mar 2026
• Accepted
  26 June 2026