October 26, 2026
Development of a recruitment and selection panel to improve the monitoring of selection processes
Ana Carolina Pereira De Araujo; Stefano Francisco Pereira Duarte
DOI: 10.22167/2675-6528-202600521
Article derived from a Course Conclusion Work (TCC), with content based on the student’s original work and adapted to the editorial format of the E&S Magazine with the support of the ResumeAI tool, an artificial intelligence solution developed by Instituto Pecege for textual synthesis and organization.
Summary
The study addressed the challenges in monitoring selection processes, arising from data fragmentation, lack of standardization, and manual report generation, which compromised visibility and decision-making. The objective was to structure an analytical solution capable of organizing, integrating, and providing information in an automated, clear, and accessible manner. The methodology adopted the PDCA cycle, with the Ishikawa Diagram for cause identification and the 5W2H plan for action planning. Operational databases were processed and standardized via Power Query, enabling the construction of a structured data model in Power BI and the development of visualizations aligned with the process needs. As a main result, the consolidation of scattered data, increased information reliability, reduction of rework, and expanded transparency and user autonomy in indicator monitoring were observed. The time to generate reports decreased from 1h30-2h to 10 minutes daily, and the update frequency changed from weekly to daily, with direct access to the dashboard. It was concluded that the developed dashboard met the study’s purpose, configuring itself as a practical, sustainable, and applicable solution for Recruitment and Selection management, strengthening data-driven decision-making.
Keywords: Process monitoring; Business Intelligence; Data management; PDCA; Decision making.
1. Introduction
In the contemporary corporate context, characterized by intense digital transformation and the growing demand for data-driven decisions, the area of recruitment and selection (R&S) assumes an increasingly strategic role in organizations. The competitiveness of the job market, the pressure for greater operational efficiency, and the need for predictability in selection processes make structured monitoring of the stages an essential requirement for aligning Human Resources with other requesting areas (Breviário et al., 2025; Peeters; Paauwe; Van De Voorde, 2020; Rasmussen; Ulrich; Ulrich, 2024). In this scenario, Business Intelligence (BI) tools have stood out for their intrinsic ability to organize, integrate, and visualize data clearly, dynamically, and strategically, offering robust management support (Microsoft, 2024; PMI, 2021).
Despite technological advancements, many R&S departments still face significant challenges related to information fragmentation, lack of standardization in databases, and difficulty in consolidating essential metrics for decision-making. In the specific case studied, the absence of a unified view of the recruitment funnel generated considerable rework for the Recruitment and Selection team. The manual preparation of weekly reports, necessary to serve managers and internal clients, consumed time, increased the probability of operational errors, and delayed the delivery of crucial business information, such as vacancy status, deadlines, and average time in each process phase.
Additionally, the limited visualization of data made it difficult to identify bottlenecks in the workflow, making tracking unintuitive and restricting the strategic potential of the Human Resources sector. More complete analyses depended on successive data consolidation rework. This operational gap, evidenced by both specialized literature and the team’s daily practice, demonstrates the urgent need for more robust analytical structures capable of transforming raw data into accessible, reliable, and up-to-date information to support decisions (Milani et al., 2020; Ghatak, 2022).
In this context, performance measurement through indicators becomes a key element for the professionalization of Recruitment and Selection, allowing the process to be driven by concrete evidence, rather than predominantly by subjective perceptions. Indicators such as average time to fill vacancies, volume of open positions, and adherence to deadlines enable greater process control, identification of inefficiencies, and support for managerial decisions. The literature indicates that the adoption of analytical practices in HR, often referred to as People Analytics, contributes to increased predictability, operational efficiency, and strategic alignment of the area with organizational objectives, reinforcing HR’s role as a strategic business partner (Milani et al., 2020; Peeters; Paauwe; Van De Voorde, 2020; Rasmussen; Ulrich; Ulrich, 2024).
The development of analytical solutions applied to the organizational context converges with approaches of continuous improvement and Quality Management, in which the structured use of data plays a central role in the control and enhancement of processes. The Plan-Do-Check-Act (PDCA) cycle, widely used in corporate environments, provides a solid methodological basis for identifying problems, implementing improvements, monitoring results, and standardizing more efficient practices (Werkema, 2012). Thus, the construction of a Business Intelligence dashboard aimed at monitoring the selection process is configured not only as a technological solution but also as a management tool, capable of supporting continuous monitoring, reducing operational variabilities, and offering greater reliability to the information used by the Recruitment and Selection team and by the managers involved.
Given this scenario, this work proposes the development of a Business Intelligence dashboard aimed at monitoring selection processes, with the objective of reducing rework, increasing information transparency, optimizing response time to internal clients, and improving decision-making in the Recruitment and Selection area.
2. Material and Methods
This study was characterized as an applied research, of a descriptive nature, with a qualitative methodological approach, focused on the development of a practical solution for an organizational problem. The empirical object was the Recruitment and Selection (R&S) process of a large company in the electricity sector, with national operations and approximately 16,000 employees. The unit of analysis focused on the R&S department, which had about nine recruiters, one supervisor, and one manager, responsible for conducting and monitoring the selection processes.
To structure the project development, the PDCA cycle methodology (Plan, Do, Check, Act) was adopted, widely used in continuous improvement and quality management processes (Werkema, 2012). This approach allowed the study to be divided into interdependent and sequential stages, ensuring coherence between the diagnosis, planning, development, evaluation, and the necessary adjustments for the construction of the Business Intelligence dashboard.
In the Planning (Plan) stage, the central problem, the objectives for building the dashboard, and the information collection criteria were defined. A meeting was held with the project’s focal people, during which the Ishikawa Diagram was applied to identify and organize the main factors hindering the monitoring of selection process data (Werkema, 2012). Complementarily, the 5W2H plan was used to guide the definition of actions foreseen in the planning.
The data collection was based on the operational spreadsheet used by the Recruitment and Selection team, updated daily by recruiters with information on the progress of selection processes. These data were subjected to organization, standardization, and structuring procedures in Power Query, in order to ensure adequate conditions for analytical modeling, according to good data preparation practices (Kimball; Ross, 2013).
In the Execution (Do) stage, data were treated and organized in Power Query, including cleaning, integration, date standardization, and textual field review. The historical database was integrated, encompassing data from selection processes registered since 2023, and continuous consolidation between historical data and updated records was automated. The team was communicated about the updates and received guidance on best practices.
With the data structure completed, the construction of the dashboard in Power BI began. The data model was established, defining the relationships between the main Recruitment and Selection table and auxiliary tables, such as companies, cities, professional levels, recruitment types, and SLA classifications. DAX measures were developed for calculating the average SLA of vacancies, performance indicators by stage, and process count metrics (Microsoft, 2024; PMI, 2021).
Filters, segmentations, and interactive parameters were configured to provide clear and intuitive navigation. Visualizations were defined by selecting charts, cards, and dashboards that best represented the relevant information, resulting in an organized and dynamic interface. SLA (Service Level Agreement) tracking by stage of the recruitment and selection process was implemented to measure the average time vacancies remained in each phase of the R&S flow.
The calculation of the average SLA per stage was performed based on the sum of the total days that vacancies remained in a given stage, divided by the number of vacancies that passed through that phase. To ensure the reliability of the indicator, the time vacancies spent in each stage was determined from the number of times each vacancy appeared in a given phase, based on daily backups performed in the Recruitment and Selection database.
In the Verification (Check) stage, the panel’s adherence to the area’s informational needs, the stability of the data flow, and the conformity of the source base to standardization rules were evaluated. The panel analysis was guided by previously defined criteria, considering the clarity of the visualizations, adherence to essential indicators, data structure coherence, and alignment with best practices in information management and visualization (PMI, 2021; Microsoft, 2024).
Finally, in the Action (Act) stage, specific adjustments were made to re-establish data flow stability and ensure the continuity of the dashboard update. Guidelines were reinforced to the Recruitment and Selection team regarding the importance of not altering file names, tabs, or columns in the operational database. Structural recovery adjustments were made in Power Query, and measures and relationships in Power BI were reviewed, including rewriting the SLA measure with validation rules and handling of null values (Werkema, 2012).
3. Results and Discussion
The initial project planning stage involved a meeting with stakeholders, in which the Ishikawa Diagram, also known as the Cause and Effect Diagram, was employed. The objective was to identify and organize the main factors hindering the visualization and adequate tracking of selection process data. This tool allowed for a structured analysis of the problem, grouping the causes into classic categories such as methods, labor, measurement, materials, machines, and environment, according to Ishikawa’s methodology (1985).
The application of the Ishikawa Diagram revealed that the problem was not restricted to a single operational point, but stemmed from structural and management failures that directly impacted the efficiency of the Recruitment and Selection process. Among the main factors identified were the lack of team training and guidance, resistance to change, absence of database standardization, recurring errors in information consolidation, and the lack of automation to ensure consistent data updates. These findings are in line with Werkema (2012), who emphasizes the importance of structured root cause identification in continuous improvement projects.
Based on the multifactorial diagnosis performed, a detailed action plan was developed using the 5W2H methodology, which transformed the identified causes into a set of practical and organized actions. This plan guided the definition of the project’s subsequent activities, ensuring an objective and targeted approach to problem-solving. The structuring of actions through 5W2H is fundamental for transforming diagnoses into concrete initiatives, especially in contexts of continuous improvement (Werkema, 2012).
The development of a detailed schedule, aligned with the 5W2H plan and the PDCA cycle, organized the project activities from September 2025 to March 2026. The planning phase focused on database analysis, initial information structuring, and prototype creation. The execution stage, which extended from October 2025 to January 2026, included adjustments to the Business Intelligence dashboard, automation implementation, and Service Level Agreement (SLA) monitoring.
The verification and results phases occurred between January and March 2026, allowing for the analysis of areas for improvement and the assessment of the solution’s implementation effects. This temporal organization ensured that the project evolved in a structured and progressive manner, with validation of intermediate deliverables, which is crucial for the control and success of complex initiatives, as pointed out by the Project Management Institute (PMI, 2021). The disciplined application of the Plan-Do-Check-Act cycle, combined with well-defined planning, increases the effectiveness of continuous improvement actions and reduces rework (Werkema, 2012).
The data processing stage in Power Query was fundamental to ensure the integrity and consistency of the information. The tool was configured to automatically copy the operational base used by recruiters and apply systematic transformations, such as formatting adjustments, date standardization, review of textual fields, and consolidation of duplicate records. These practices are essential for the reliability of Business Intelligence solutions, according to the recommendations of Kimball and Ross (2013).
In addition, the historical base of selection processes, recorded since 2023, was also processed and standardized in Power Query. This integration allowed for continuous consolidation between historical data and updated records, expanding the volume of information and the analytical depth of the monitoring. The integration of historical and operational data is a crucial aspect for consistent analyses and for supporting managerial decision-making (PMI, 2021; Milani et al., 2020).
With the data structure consolidated, the construction of the dashboard in Power BI began. Firstly, the data model was established, defining the relationships between the main Recruitment and Selection table and auxiliary tables, such as companies, cities, professional levels, recruitment types, and SLA classifications. This model ensured referential integrity and consistency of the analyses, being a crucial step in building a robust BI system.
After modeling, DAX measures were developed to calculate the average SLA of vacancies, performance indicators by stage of the selection process, and process count metrics, enabling structured monitoring of the recruitment funnel. Subsequently, interactive filters, slicers, and parameters were configured, providing clear and intuitive navigation for users, resulting in an organized and dynamic interface aligned with the sector’s needs.
During the execution phase, SLA (Service Level Agreement) monitoring was implemented per stage of the selection process, with the objective of measuring the average time vacancies remained in each phase of the Recruitment and Selection flow. To enable this indicator, historical records of vacancy movements were adopted through daily backups of the operational database, ensuring the traceability of status changes over time.
The consolidation of this data in Power Query allowed for the precise determination of the average time vacancies remained in each stage, offering a detailed view of the funnel’s behavior and facilitating the identification of operational bottlenecks. The implementation of this control significantly enhanced the robustness of the analytical dashboard by reducing distortions and enabling consistent temporal analyses of performance indicators, which is fundamental for strategic decision-making (Kimball and Ross, 2013; Rasmussen et al., 2024).
In the Check stage of the PDCA cycle, operational weaknesses related to data governance were identified. Interruptions in Power Query and Power BI were observed due to changes in the source file name, which compromised references and temporarily prevented automatic updates. These findings reinforce the sensitivity of Business Intelligence solutions to uncontrolled changes in the structure of source data, as highlighted by Kimball and Ross (2013).
Additionally, the absence of a record of the number of interviewed candidates was identified, which made it impossible to calculate the conversion rate between interviewed and hired individuals. In the SLA monitoring by stage, it was verified that part of the team did not update the status of vacancies daily, generating distortions in the calculation of time per stage, including cases of negative SLA due to invalid dates. These results reinforce the dependence between the quality of data input and the reliability of indicators (Milani et al., 2020; Rasmussen et al., 2024).
Based on the identified inconsistencies, punctual adjustments were made to re-establish data flow stability and ensure the continuity of the dashboard update. Guidelines were reinforced to the Recruitment and Selection team regarding the importance of not altering file names, tabs, or columns in the operational database, preventing operational failures such as broken connections and temporary loss of information in the analytical model. These preventive actions are crucial for maintaining a stable analytical environment.
In the technical scope, structural recovery adjustments were made in Power Query, reestablishing removed columns and revalidating compromised transformation steps. In Power BI, the SLA measure was rewritten to incorporate validation rules, null value handling, and safeguards against illogical results, such as negative SLAs. These actions allowed for the recovery of the solution’s full functionality and reinforced preventive maintenance practices, preparing the analytical environment for continuous improvements, in line with the principles of continuous improvement (Werkema, 2012).
General Discussion of Results
The implementation of the Business Intelligence dashboard allowed the consolidation of data that was previously dispersed, resulting in a structured, integrated, and automatically updated database. The application of diagnostic and planning tools, such as the Ishikawa Diagram and the 5W2H plan, enabled targeted action on the factors limiting the efficiency of the selection process. The standardization of the database, combined with the use of Power Query and the construction of the analytical model in Power BI, contributed to the availability of more reliable, clear, and accessible information for the Recruitment and Selection team.
Qualitatively, the results showed improvements in bottleneck identification, recruitment funnel visibility, SLA accuracy per stage, and the possibility of comparative analyses by area, job type, or process phase. The developed solution establishes a solid foundation for future evolutions, such as the expansion of strategic indicators and the enhancement of automations, contributing to the increased analytical maturity of the Recruitment and Selection area.
The quantitative results reinforce the impact of the implemented solution, showing a significant reduction in the time required for generating the management report. Previously, the preparation of the weekly report demanded between 1h30 and 2h, involving the manual consolidation of the operational spreadsheet and sending information via email. With the implementation of the Power BI dashboard, the total update time became approximately 10 minutes per day, already including basic data integrity checks. This reduction represents a significant operational gain, freeing up the team’s time for analytical and strategic activities (Milani et al., 2020).
Another relevant aspect is the frequency of information updates, which evolved from a weekly model to daily monitoring. This change significantly increased the visibility of the recruitment funnel and enabled continuous monitoring of job openings. Frequent access to updated indicators strengthens the predictability of Human Resources processes and contributes to decisions more aligned with organizational needs (Peeters; Paauwe; Van De Voorde, 2020).
The increased autonomy of Business Partners in accessing information is also highlighted, which no longer depends on the manual sending of reports by email but occurs through direct access to the dashboard. This change contributed to greater process transparency and information democratization, expanding the strategic role of HR by allowing different stakeholders to monitor indicators in real-time (Rasmussen et al., 2024). Information reliability also advanced significantly, moving from vulnerable to manual errors to high, due to the standardization of the database and automation of the workflow.
Thus, the results obtained confirm that the implementation of the Business Intelligence panel promoted consistent gains in operational efficiency, reduction of rework, and greater transparency in monitoring selection processes. The increase in informational agility and communication flow between the Recruitment and Selection team and Business Partners evidences the advancement of the area’s analytical maturity, contributing to more efficient and data-driven decision-making, fully meeting the study’s objective.
4. Conclusion
The present study sought to structure an analytical solution to organize, integrate, and provide information in an automated, clear, and accessible manner, aiming to improve the monitoring of selection processes and decision-making in the Recruitment and Selection area. It was verified that the implementation of the Business Intelligence dashboard, built using Power Query for data processing and standardization and Power BI for modeling and visualization, allowed the consolidation of data that was previously dispersed. A significant increase in information reliability, an expressive reduction in rework, and an expansion of user transparency and autonomy in indicator monitoring were observed. Quantitatively, the time to generate management reports decreased from one hour and thirty minutes to two hours to approximately ten minutes daily, and the update frequency changed from weekly to daily, with direct access to the dashboard by Business Partners. The developed solution is configured as a practical and sustainable contribution to the management of Recruitment and Selection, strengthening data-driven decision-making and driving the area’s analytical maturity.
Despite the advances, operational limitations related to data governance were identified, such as the system’s sensitivity to changes in source file names and the inconsistency in the daily update of vacancy status by the team, which generated distortions in the calculation of time per stage. Additionally, the absence of records for the number of interviewed candidates made it impossible to calculate the conversion rate. For future studies, it is suggested to expand strategic indicators, improve automations, and incorporate additional data for more comprehensive analyses, such as the conversion rate between stages. The developed dashboard establishes a robust foundation for continuous monitoring and process improvement, consolidating the strategic role of the Human Resources area through more efficient and data-driven management.
Bibliographic References
Breviário, Á. G. do; Rocha, L. A. C.; Rebello, F. A. S.; Vidal, R. L. da S.; Oliveira Filho, A. de S. 2025. Recrutamento e seleção no século XXI: superando barreiras digitais, critérios tradicionais e potenciais não convencionais. Revista Eletrônica Multidisciplinar de Investigação Científica 4(21): e1181825.
Ghatak, R. 2022. People analytics: data to decisions. Springer, Singapore
Ishikawa, 1985 [Referência completa não encontrada no documento original]
Kimball, R.; Ross, M. 2013. The data warehouse toolkit: the definitive guide to dimensional modeling. 3. ed. Hoboken: Wiley.
Microsoft. 2024. Human resources sample: take a tour – Power BI. Microsoft Learn. Disponível em: <https://learn.microsoft.com/en-us/power-bi/create-reports/sample-human-resources>. Acesso em: 16 jan. 2026
Milani, A. M. P.; Soares, J. A.; Andrade, G. L.; et al. 2020. Visualização de dados. SAGAH, Porto Alegre, RS, Brasil
Peeters, T.; Paauwe, J.; Van De Voorde, K. 2020. People analytics effectiveness: developing a framework. Journal of Organizational Effectiveness: People and Performance 7(2): 203–219.
Project Management Institute [PMI]. 2021. Um guia para o conhecimento em gerenciamento de projetos (Guia PMBOK®). 7. ed. PMI, Pennsylvania, USA
Rasmussen, T. H.; Ulrich, M.; Ulrich, D. 2024. Moving people analytics from insight to impact. Human Resource Development Review 23(1): 11-29.
Werkema, C. 2012. Métodos PDCA e DMAIC e suas ferramentas analíticas. GEN/Atlas, Rio de Janeiro, RJ, Brasil
Article originating from the Final Course Project of the Specialization in Project Management of the MBA USP/Esalq
To learn more about the course, click here and access the MBX Academy platform