**5. Discussion**

The descriptive analysis allowed us to identify that the SMEs studied perceived that their managemen<sup>t</sup> control tools were well defined (69%). In addition, decision-making based on rational and financial quantitative information was employed, which focuses organizational culture towards results. Regarding performance, the financial area shows improvement, with 64% of answers positively asserting that utility increased after employing managemen<sup>t</sup> control tools for two years. Furthermore, 57% consider that their liquidity level improved, and slightly less than half report that their debt levels lowered. Regarding non-financial performance, 65% report improvements to the organizational environment after employing the tools, while 58.2% claim to sense better control in decision-making processes.

As a business grows, the use of control tools increases. A clear example can be seen in the beliefs lever values that spiked up to 54.8% and 86% in using strategic objectives to define strategy among small and medium-sized companies, respectively. It must be noted that the interactive lever clearly shows strategic surveillance strategies in the competition, with 58.7% of small companies and 70.3% of medium-sized companies practicing this activity.

Interestingly, the boundary lever is the least used among the SMEs that employ it. The values are similar to the previous results, with 67% coming from small companies and 84.5% from medium-sized companies. In retrospect, the diagnostics lever has clearer differences between small and medium-sized companies in costing system (56% vs. 69%), state of results (59.6% vs. 76%), and use of budget (68.4% vs. 86.2%).

The analysis contributes to what is highlighted in the literature regarding the structural weaknesses to create managemen<sup>t</sup> strategies for medium and long-term projections due to a scarcity of tools. In addition, managemen<sup>t</sup> of administration, finances, accounting, and operations is very informal and is done in an intuitive manner where few managemen<sup>t</sup> control tools are employed in decision-making [15]. Furthermore, there is evidence of the low use of managemen<sup>t</sup> control tools by small companies, which does improve when viewing medium-sized companies in agreemen<sup>t</sup> with the studies by Frost [20].

Nonetheless, we can perceive that using managemen<sup>t</sup> tools does improve performance without substantial difference between financial and non-financial performance. This reinforces the literature that states that the use of managemen<sup>t</sup> control tools has a positive impact on organizational performance [6,10,22]. In regard to the measuring scale employed in this study, it had been validated as it returned a high level of internal data consistency through the Cronbach Alpha [44], satisfactory statistical values with the Kaiser–Meyer– Olkin method, and acceptable numbers in Bartlett's test. Lastly, the factorial analysis identified factors that recommended that we eliminate certain variables, keeping the original theoretical model for the one that was finally used. This separated beliefs into strategy and planning and separated the diagnostics lever into financial and non-financial diagnostics. These were analyzed by means of fuzzy logic, as indicated in Figure 6.

#### *Fuzzy Inference System (FIS) for Management Control Tool Levers*

The fuzzy inference system revealed that all levers were used at 'medium' capacity by SMEs, see Table 3. The least employed lever was "Boundary", which includes the use of formal contracts, profile descriptions, manuals, organizational structure, and role hierarchy. Results show that the degree of membership was −7 without significant change between small and medium-sized companies (see Table 3). This indicator remained stable as the other variables of the lever remained in 'medium' use, but with a negative trend. In retrospect, the other levers have positive use led by interactive control, followed by diagnostics control, and lastly by beliefs. Performance is valued as 'medium' due to the median use of managemen<sup>t</sup> control tools, grouped in control levers (Simons, 1995) by SMEs based on the literature.

We recommend training mainly the boundary lever, as it is compared to the other levers (see Table 3), as also shown in the descriptive analysis. As mentioned in the literature, it is recommended that entrepreneurs and businesspersons are trained on how to use managemen<sup>t</sup> control tools. It is recommended that managemen<sup>t</sup> control tools be provided to small companies. This would increase the degree of membership for medium-sized companies in all levers, including the diagnostics group, but excluding the boundary group, which remains low in all SMEs. The fuzzy inference system also revealed that the degree of membership for the beliefs lever is 'medium', which is of concern in small companies (FIS = 6.0376). Due to its positive relationship with level of performance, it is recommended to train SMEs in these control tools, as different variations (of use) may imply a highly unstable variable behavior. In addition, the fuzzy inference system explains similar results found in the descriptive analysis, where the least used lever by SMEs is "Boundary", not showing differences in degrees of membership regarding financial and non-financial performance.

Lastly, the study suggests that managemen<sup>t</sup> control tools in SMEs have a mid-level of use or degree of membership. "Boundary" was the least used, followed by "Beliefs", "Diagnostics", and "Interactive" (as the most used). This may be because SMEs follow horizontal and organic structures, which causes information to flow and generate team collaboration for decision-making. In retrospect, it is necessary to implement training in managemen<sup>t</sup> control for businesspersons that lead SMEs, as it could increase the company's performance and keep them sustainable over time.
