Executive Summary

Isis Penelope Sumarelli Albuquerque Huang; Vítor Melão Cassânego

Summary prepared by the ResumeAI tool, an artificial intelligence solution developed by Instituto Pecege focused on synthesis and writing.

The growing complexity of agri-food supply chains, driven by globalization, has generated an intensified demand for transparency, security, and traceability in food systems. The integrity and authenticity of products are crucial, making the implementation of specific and reliable protocols for traceability a priority in the industry, as pointed out by Rossi et al. (2025). This scenario is aggravated by the growing consumer demand for greater clarity on food quality and safety, as well as for premium offerings, as highlighted by Charlebois (2024).

In this context, the supplier qualification process emerged as a fundamental strategic pillar to ensure compliance with regulatory standards, mitigate risks, and guarantee the quality and safety of inputs used in the production of food for human and animal consumption. The robustness of this process is vital for the agile acquisition of crucial information, which supports strategic and operational decision-making throughout the organization’s value chain.

The risk scenario in the supply chain was dramatically evidenced in 2020, when the detection of ethylene oxide in sesame seeds imported from India to Europe triggered a global alert. This incident, as reported by Anyogu et al. (2024), exposed consumers to contaminants and underscored the urgency of strengthening control mechanisms and traceability of information on the origin and processing of ingredients, especially those of plant origin, which are particularly vulnerable to such contamination.

In a large multinational corporation, the supplier qualification process for ingredients faced significant challenges. The global questionnaires used were predominantly composed of open-ended and discursive questions, which hindered the collection and analysis of structured data. This manual and isolated approach prevented effective comparative analysis and reduced the ability to implement corrective and preventive actions on a global scale, compromising the strategic vision of the supply chain.

The fragility of the data collection and analysis process was highlighted after the 2020 episode, driving a project to review and standardize global questionnaires. The development of a new model, with about 85% multiple-choice questions and the use of the Qualtrics platform, brought improvements in the quality and speed of the collected data. However, the management of this data still represented a challenge, as each region used a distinct link in Qualtrics, preventing an integrated and real-time visualization of risks and trends.

The central problem, therefore, was not limited to mere data collection, but to the absence of effective technological integration and a robust, predictive analytical process. The organization operated reactively, responding to deviations only after they occurred. This reinforced the urgent need to transition to a more integrated and intelligent supplier qualification model, capable of mitigating risks, strengthening regulatory compliance, and supporting strategic decisions across the global supply chain.

The analysis of the situation revealed that the supplier qualification process, in place since 2012, presented critical limitations in data collection, interpretation, and consolidation. The discursive nature of the questionnaires resulted in manual and isolated analyses, fragmenting information and hindering traceability. This fragmentation prevented the visualization of patterns and agile reaction to emerging risks in the supply chain, compromising operational efficiency and safety.

The complexity of the global ingredient supply chain, with suppliers in various countries subject to local legislation and different levels of maturity in their production and quality processes, aggravated the problem. The absence of integrated global dashboards prevented a clear and aligned view of the situation across all regions, reducing the ability to identify risks and opportunities in real-time, which was crucial for global management.

The relevance of the problem was accentuated in 2021, with the increase in international alerts regarding ingredient contamination. Cases such as the presence of ethylene oxide residues in seeds and spices led to hundreds of notifications registered in the European Rapid Alert System for Food and Feed (RASFF) between 2020 and 2025, totaling 1003 occurrences, the majority originating from India (RASFF, 2025). These occurrences pressured regulatory authorities, resulting in specific regulations in the European Union and Brazil.

The failure to identify and treat these contaminants in a timely manner can lead to severe consequences, including product recalls, substantial financial losses, and irreparable damage to the reputation of the companies involved. A striking example was the recall of Häagen-Dazs vanilla ice cream in 2022, due to the detection of 2-chloroethanol residues. The National Health Surveillance Agency (ANVISA) confirmed the voluntary recall in Brazil (Agência Brasil, 2022; ANVISA, 2022), highlighting the criticality of the absence of integrated monitoring in global supply chains.

The analysis of the root causes of the problem, using a logic similar to the “5 whys” method, revealed that the difficulty was not restricted to the format of the questionnaires, but also to the information management model. Although open-ended questionnaires were applied globally, the interpretation varied between regions, and the absence of a clear global guideline and integrated technological guidance allowed each region to conduct the process autonomously, resulting in a decentralized and ineffective dynamic for the global strategy.

Recent studies corroborate that even organizations with large volumes of data face challenges when relying exclusively on numbers without considering context, which can lead to biased decisions (Westover, 2024). Research by Alimi et al. (2024) identified recurring problems such as low data quality, system fragmentation, lack of skilled professionals, and internal resistance to change. Thus, the need to develop a solid analytical culture, where employees interpret and apply data with meaning, has become imperative (Tobin, 2024; Sheehan, 2025).

Faced with this diagnosis, the project aimed to develop a globally aligned database, using the Qualtrics platform integrated with Business Intelligence (BI) tools, such as Power BI. The objective was to enable the collection and consolidation of questionnaire information, generating performance and risk indicators, as well as predictive models to anticipate failures and opportunities. This would reinforce standardization and alignment with the organization’s global strategic goals, transforming the reactive stance into a proactive and intelligent approach.

To address the limitations of the global supplier qualification process, two main lines of work were identified. The first alternative proposed the development of an integrated alert database, inspired by the European RASFF system, connected to governmental platforms and other external monitoring sources. The second alternative consisted of reviewing and standardizing the global supplier qualification questionnaires for ingredients, replacing the discursive model with a more structured and operational format on the Qualtrics platform.

The first alternative, focused on the integrated alert base, aimed to consolidate information from national bodies such as ANVISA and MAPA, and international ones like RASFF, FDA, and EFSA. This solution contemplated the automatic capture of alerts and regulatory updates via APIs, allowing for systematic cross-referencing of alerts with supplier and raw material registrations. The scope included the structuring of a detailed history of occurrences, categorized by ingredient, country of origin, and severity, providing a centralized view of supply chain vulnerabilities.

The expected deliverables for the first alternative included a centralized alert platform, an API integration module, a structured and categorized database, standardized operational reports, and a system for mapping potentially impacted suppliers. The execution depended on specialized human resources, such as developers, risk analysts, and regional teams, in addition to significant technological resources, such as internal servers, integration systems, API licenses, security layers, firewalls, and cloud database infrastructure.

From a financial point of view, the alert base alternative would involve a moderate to high initial investment, with recurring maintenance costs and eventual hiring of external specialists. The estimated execution timeframe would be six to twelve months, which proved to be longer than the period available for the project. Although its strategic potential was high, the technological complexity, dependence on multiple external databases, and the need for continuous data curation, in addition to the volume of resources incompatible with the deadline, made this alternative unfeasible for the current project cycle.

The second alternative, which was the recommended solution, focused on the complete revision and standardization of the global ingredient supplier qualification questionnaires. The objective was to transform the questionnaires, which were entirely discursive, into a single, more structured and standardized model, with approximately 85% multiple-choice questions, integrated into the Qualtrics platform. This approach aimed to optimize data collection and analysis, making them more comparable and useful for decision-making.

The scope of this alternative involved the technical review of the questionnaire content, aligning it with global and regional regulatory requirements. Essential fields for risk assessment were defined, such as country of origin, harvesting process, laboratory analyses, certifications, and controls applied to the ingredient. The standardization of interpretive criteria among the company’s regions was a crucial point, ensuring consistency and uniformity in the global supplier assessment.

The great advantage of this alternative was the exclusive use of technological and human resources already available within the organization. There was no need to acquire new systems, additional licenses, or hire extra personnel. The resources included regional Quality and Supply Chain representatives, a Qualtrics specialist, and punctual support from the IT team, in addition to corporate Excel spreadsheets and the company’s internal network and collaboration environment, such as SharePoint.

The expected deliverables for the second alternative included the standardized and operational global questionnaire in Qualtrics, a global data manual for concept standardization, and an Excel extraction “pipeline” for regional and global analyses. Filling guides for suppliers, an internal manual for the evaluation teams, and training materials were also developed. These elements allowed for the uniform structuring of the qualification flow, ensuring comparability and greater risk visibility.

From a financial point of view, the incremental investment was minimal, given the absence of the need to acquire additional systems or licenses. The estimated execution timeframe was two to four months, which proved compatible with the project’s constraints. This alternative proved adequate for the context, as it transformed a heterogeneous form into a structured and standardized instrument, reducing response variability and increasing data reliability.

Studies on traceability and standardization of information, such as those by Rossi et al. (2025) and Charlebois et al. (2024), reinforce that processes with clear structures and uniform protocols improve information transparency and quality. The adoption of the structured questionnaire brought internal benefits, such as greater agility in information processing and improved alignment between regions, strengthening global governance and the organization’s data culture.

The cost-benefit analysis reinforced the choice of the second alternative. Among the tangible benefits, the reduction in the analysis time of collected information, the increase in data reliability, the decrease in interpretation errors, and the improvement in comparability between regions stood out. Intangible benefits included the strengthening of global governance, greater transparency for user areas, stimulation of a data culture, and advancement in the organization’s analytical maturity, essential elements for effective risk management.

In contrast, the absence of action would keep the process fragmented, dependent on manual analyses, vulnerable to inconsistencies, and susceptible to risks already evidenced in international episodes of ingredient contamination. This could result in potential economic losses and significant reputational impacts, compromising the competitiveness and safety of the company’s supply chain, according to the scenario analysis performed.

To support the final decision, a multi-criteria evaluation matrix based on the “Multi-Criteria Decision Making” (MCDM) approach was used, as described by Velasquez and Hester (2013). This methodology increased consistency and transparency in the decision-making process, considering multiple factors simultaneously. The matrix evaluated criteria such as technical feasibility, complexity, deadline, cost, technological dependence, risk of delay, immediate impact, and strategic impact for both proposed alternatives.

The comparison revealed that the alternative of reviewing and standardizing the global questionnaires performed better on most criteria, standing out for its high technical feasibility, speed of implementation, and alignment with existing resources. In contrast, the integrated alert base alternative, despite its high long-term strategic potential, involved higher costs, greater complexity, high technological dependence, and the need for resources not foreseen in the initial scope, making it less suitable for the present moment.

Based on this multicriteria analysis, the review and standardization of the global supplier qualification questionnaires for ingredients was the recommended priority solution. For its implementation, a high-level plan composed of five stages was structured. The first stage involved the technical review of the questionnaire content, aligning it with regulatory requirements and regional and global needs, conducted by the project leader in conjunction with representatives from each region.

The second stage consisted of the construction and testing of the new questionnaire on the Qualtrics platform, led by the project leader, to ensure the correct parameterization, logic, and functioning of the instrument. Subsequently, the third stage focused on validation with regional representatives, including pilot tests with suppliers, executed by the project leader and regional representatives, ensuring the applicability and effectiveness of the new model in different operational contexts.

The fourth stage involved the integration of the collected data with Business Intelligence tools, enabling global and regional comparative analyses. This integration was carried out by the IT team in conjunction with the project leader, aiming to transform raw data into actionable strategic information. Finally, the fifth stage involved training the user teams and global implementation, conducted by the project leader to ensure uniform and effective adoption of the new process in all company regions.

The planned deliverables included the new standardized questionnaire integrated into the Qualtrics platform, allowing its structured and centralized application by regional teams. A global data dictionary was developed for concept standardization, a “pipeline” for standardized extraction of responses for regional and global analysis, preliminary analytical reports, and support materials for the use and interpretation of results, such as guides and internal instructions, facilitating the transition and operation of the new system.

The project’s main dependency relationships involved consolidating the questionnaire on the Qualtrics platform before integration with BI, and regional validation before global implementation. Resources used focused on existing technological infrastructure, hours allocated to regional and global teams, and the support of the IT and Analytics team, optimizing the use of company assets and minimizing the need for additional investments.

The project’s functional structure, detailing each stage, its responsible parties, dependencies, and deliverables, served as a clear operational guide for implementation. This approach ensured clarity in the division of responsibilities, predictability in the progress of activities, and adequate control of dependencies between stages. Thus, the selected alternative met the deadline constraints, mitigated the main identified risks, and established the foundations for future developments, including the possibility of developing, in a later stage, the integrated alert base inspired by the RASFF model.

The evaluation of the implemented solution was structured based on performance indicators aligned with the project’s strategic objectives: increasing the efficiency of the qualification process, expanding risk visibility, and strengthening regional governance. For each objective, measurable metrics, data collection methodology, analysis periodicity, and responsible parties for monitoring were established, ensuring continuous and effective monitoring of the achieved results.

On the operational efficiency axis, the defined indicator was the average supplier response time to qualification questionnaires. Historically, this period was 15 business days. The established goal was to reduce this time by at least 30% in the first cycle after implementation. After implementation, data extracted directly from the Qualtrics platform and consolidated into standardized Excel spreadsheets showed that the average period was reduced to ten business days, exceeding the planned goal and evidencing greater clarity of the questions, reduction of rework, and greater agility in completion by the suppliers.

On the risk visibility axis, indicators included the completeness of critical information, such as country of origin, harvest type, current certifications, laboratory analyses performed, and controls applied, as well as the ability to segment suppliers by risk level. The goal was to achieve 95% completion of critical fields and enable the structured classification of all high-impact raw material suppliers. Monitoring demonstrated consistent progress in information standardization and allowed for comparative analyses between regions, in addition to the preventive identification of weaknesses that were not previously evidenced in a structured manner.

On the regional governance axis, the established indicators were the regularity of the analysis of received responses and the processing time for any documentary inconsistencies. The goal was to conduct weekly reviews of submitted responses and address critical deviations within five business days. The analyses were carried out by regional ingredient category managers, with consolidated reporting to the global team whenever relevant risks were identified, ensuring a rapid and coordinated response to any issues.

The monitoring methodology was based on the weekly extraction of data from Qualtrics, consolidation into structured spreadsheets, and periodic review of metrics in operational meetings. Although automated integration with Business Intelligence tools was planned for a later stage, there was insufficient time for its implementation in the analyzed period. However, the adopted model ensured traceability, transparency, and consistency in information analysis, even without full automation.

The continuous monitoring plan included weekly assessments at the regional level, consolidated monthly reviews at the global level, and semi-annual re-evaluations of strategic indicators. This model allowed for verification of whether the solution delivered the expected results and provided subsidies for future adjustments and evolutions in the supplier qualification process, ensuring continuous improvement and adaptation to new market and regulatory demands.

The evaluation confirmed that the implemented solution generated measurable efficiency gains, expanded risk predictability, and strengthened process governance. These results established solid foundations for future stages of analytical integration and digital evolution, positioning the organization for more proactive and strategic management of its supply chain, mitigating vulnerabilities and increasing operational resilience in an increasingly demanding business environment.

The analysis developed showed that the global supplier qualification process presented relevant limitations, especially related to the fragmentation of information, the low standardization of responses, and the difficulty of transforming the collected data into strategic inputs for decision-making. The absence of an integrated analytical structure reduced risk visibility and generated operational rework, impacting the efficiency and predictability of the supply chain, a critical problem for global food security.

Given this scenario, the recommendation was to review and standardize the global supplier qualification questionnaires on the Qualtrics platform, replacing the predominantly discursive model with a structured, comparable, and indicator-oriented format. The solution prioritized the use of tools already available within the organization, demanded moderate financial investment, and concentrated efforts mainly on technical hours for review, parameterization, and training of the teams involved, optimizing existing resources.

The results obtained confirmed tangible gains, such as the reduction in average supplier response time, the decrease in rework in analyses, and greater regularity in regional monitoring. Furthermore, they generated relevant intangible benefits, including greater transparency in the decision-making process, strengthening of internal governance, increased information reliability, and greater maturity in supply chain risk management, crucial elements for the organization’s long-term sustainability and competitiveness.

It was concluded that the implemented solution not only met the established strategic objectives but also created solid foundations for future analytical evolutions, including eventual integration with BI tools and external monitoring sources. Given the results achieved, operational feasibility, and alignment with organizational strategy, the consolidation and expansion of the structured qualification model were configured as the most adequate decision to sustain continuous improvement and proactive risk mitigation in the global business environment, ensuring product safety and integrity.

Bibliographic References:

AGÊNCIA BRASIL. 2022. General Mills recalls Häagen-Dazs vanilla ice cream in Brazil. Disponível em: https://agenciabrasil.ebc.com.br/en/geral/noticia/2022-07/general-mills-recalls-haagen-dazs-vanilla-ice-cream-brazil. Acesso em: 29 ago. 2025.

AGÊNCIA NACIONAL DE VIGILÂNCIA SANITÁRIA [ANVISA]. 2022. Recolhimento de lotes de sorvetes da marca Häagen-Dazs. Disponível em: https://www.gov.br/anvisa/pt-br/assuntos/noticias-anvisa/2022/recolhimento-de-lotes-de-sorvetes-da-marca-haagen-dazs. Acesso em: 29 ago. 2025.

Alimi, A.; Grace, C.; Oladele, S.; Mia, J. 2024. Data-Driven Decision-Making: Challenges and Opportunities in Business Analytics. RESEARCHGATE. Disponível em: https://www.researchgate.net/publication/387365368 Data-Driven Decision-Making Challenges and Opportunities in Business Analytic. Acesso em: 27 ago. 2025.

Anyogu, A.; Somorin, Y. M.; Oladipo, A.O.; Raheem, S. 2024. Food safety issues associated with sesame seed value chains: Current status and future perspectives. HELIYON 10(16): e36347. Disponível em: https://doi.org/10.1016/j.heliyon.2024.e36347. Acesso em 27 ago. 2025.

Charlebois, S.; Latif, N.; Ilahi, I.; Sarker, B.; Music, J.; Vezeau, J. 2024. Digital Traceability in Agri-Food Supply Chains: A Comparative Analysis of OECD Member Countries. FOODS 13(7):1075. Disponível em: https://doi.org/10.3390/foods13071075. Acesso em: 25 ago. 2025.

European Comission [EC]. 2025. Rapid Alert System for Food and Feed – RASFF. Disponível em: https://webgate.ec.europa.eu/rasff-window/screen/search. Acesso em: 29 ago. 2025.

Rossi, S.; Gemma, S.; Borghini, F.; Perini, M.; Butini, S. Carullo, G.; Campiani, G. 2025. Agri-food traceability today: Advancing innovation towards efficiency, sustainability, ethical sourcing, and safety in food supply chains. TRENDS IN FOOD SCIENCE & TECHNOLOGY 163: 105154. Disponível em: https://doi.org/10.1016/j.tifs.2025.105154. Acesso em 25 ago. 2025.

Sheehan, T. 2025. Making better decisions with data literacy. EVINACT. Disponível em: https://www.evinact.com/articles/making-better-decisions-with-data-literacy Acesso em: 1 set. 2025.

Tobin, D. 2024. Why Data Literacy is Essential for a Data-Driven Future. INTEGRATE.IO. Disponível em: https://www.integrate.io/blog/importance-of-data-literacy. Acesso em: 1 set. 2025.

Velasquez, M; Hester, P. T. 2013. An analysis of multi-criteria decision making methods. RESEARCH GATE. Disponível em: https://www.researchgate.net/publication/275960103 An analysis of multi-criteria decision making methods. Acesso em: 15 nov 2025.

Westover, J. H. 2024. When Analysis Outruns Wisdom: Potential Pitfalls of Data-Driven Decision-Making. HUMAN CAPITAL INNOVATIONS. Disponível em: https://www.innovativehumancapital.com/article/when-analysis-outruns-wisdom-potential-pitfalls-of-data-driven-decision-making. Acesso em: 1 set. 2025.

Executive summary from the Final Course Work of the Specialization in Executive Leadership and Management of the MBA USP/Esalq

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