October 26, 2026
Institutionalization of the Kanban method in the management of multiple strategic projects in a technology startup
Maria Naomi Pereira Wagatsuma; Leticia Franceschini Rodrigues
DOI: 10.22167/2675-6528-202601196
Article prepared by the ResumeAI tool, an artificial intelligence solution developed by the Pecege Institute focused on synthesis and writing.
Summary
The institutionalization of the Kanban method in the management of multiple strategic projects in a technology startup was analyzed, motivated by the absence of a structured monitoring model and the low executive visibility of the portfolio. The study aimed to propose, apply, and evaluate a visual governance system capable of improving workflow predictability and decision-making capacity. The research was characterized as applied, with a qualitative approach and action-research design, conducted between July 2025 and March 2026. The intervention consisted of implementing a two-level visual portfolio system in the Businessmap software, with definition of workflow, work-in-progress limits, service classes, and monitoring rituals. The results indicated an improvement in workflow organization, with increased operational consistency and reduced typical variability, as well as expanded executive visibility through the centralization and explicitation of portfolio information. Limitations included the short analysis period after implementation, dependence on update discipline, and unequal adoption across areas. It was concluded that the institutionalization of Kanban contributed to structuring the workflow and supporting decision-making, even though its maturity depends on the continuous evolution of system use and governance.
Keywords: Organizational Agility; Portfolio Management; Visual Management; Kanban; WIP.
1. Introduction
The management of multiple strategic projects in fast-growing organizations, such as technology startups, imposes relevant challenges of coordination, prioritization, and execution monitoring. In environments characterized by high demand variability, interdependency between areas, and capacity limitations, the absence of structured management mechanisms compromises workflow fluidity and reduces the organization’s ability to sustain its strategic priorities.
From this perspective, the predictability of the flow and executive visibility become essential elements for decision-making and for aligning strategy with execution. When the monitoring of initiatives occurs in a fragmented manner, with low standardization of criteria and dispersed information, the risks of delays, priority conflicts, and inefficient resource allocation increase.
According to Kerzner (2018), project management maturity is associated with the coherence between organizational objectives and the ability to execute them in a coordinated, controlled manner, aligned with the strategy. Information fragmentation is one of the main obstacles to this maturity. In this context, the problem is not limited to delivery delays but also involves losses resulting from the low quality of managerial decisions. Reinertsen (2009) argues that, in systems with invisible queues and excessive work in progress, the cost of delay intensifies, harming value generation and organizational response to competing demands.
In response to this type of context, the Kanban method presents itself as a pertinent approach for managing multiple demands. According to Anderson (2010), Kanban allows for visualizing work, making flow policies explicit, and limiting Work in Progress (WIP), fostering greater balance between demand and capacity. In this logic, Kanban should not be understood merely as a visual board, but as a management system guided by explicit rules, blocking criteria, and follow-up rituals.
Berkun (2008) complements this perspective by arguing that clarity in communication and early visibility of problems are decisive factors for the effective conduct of projects in dynamic and uncertain environments. The institutionalization of its practices, therefore, aims to improve responsiveness and operational efficiency, contributing to more robust and adaptable management.
The justification for this study lies in the fragility observed in the simultaneous monitoring of multiple strategic projects in the analyzed organization. The conduction of initiatives through manual controls and dispersed information hindered the integrated visualization of the flow, compromised prioritization, increased team overload, and reduced executive visibility into the actual progress of the projects. Thus, not only an operational monitoring limitation was identified, but also a managerial gap associated with the absence of an institutionalized management system that organized the workflow in an explicit, stable, and shared manner among those involved.
Starting from the proposition that the institutionalization of the Kanban method, through the explicit definition of flow, the limitation of work in progress, and the adoption of monitoring rituals, contributed to increasing workflow predictability and executive visibility of strategic projects, this study sought to answer the following research question: in what way did the institutionalization of the Kanban method contribute to the predictability of flow and to the executive visibility of multiple strategic projects in a technology startup? Thus, the objective of this work was to propose, apply, and analyze the institutionalization of the Kanban method in the management of multiple strategic projects in a technology startup, with an emphasis on workflow predictability and executive visibility.
2. Material and Methods
This study was characterized as an applied research, with a qualitative approach and descriptive nature. The research-action design with intervention was adopted, as recommended by Thiollent (2011), allowing the researcher to act directly in the organizational context to diagnose failures, intervene in the management system, and monitor the initial effects of the change.
The research was conducted in a technology startup located in São Paulo, SP. The temporal scope of the investigation was divided into three moments: the period before the institutionalization of the Kanban method (July to December 2025), the implementation phase (starting in January 2026), and the analysis of the initial effects of the intervention (January to March 2026), focusing on the stabilization of the new system.
Two complementary units of analysis were established. At the strategic level, the 20 active strategic projects within the portfolio framework were analyzed to assess executive visibility and distribution. At the operational level, the initiatives, deliverables, and tasks linked to these projects were used to assess workflow predictability. The researcher, the leaders responsible for portfolio governance, and the representatives of the involved areas participated in the implementation.
The empirical base was composed of primary and secondary data extracted from the Businessmap software, including card movement histories, entry and exit dates, time spent per stage, blocking records, service classes, responsible parties, and work-in-progress limits. Complementarily, organizational records such as schedules, meeting minutes, and participant observation notes were analyzed. The base was debugged to prioritize records linked to strategic projects.
The intervention was operationalized through the structuring of a two-level visual governance system in Businessmap. The portfolio board focused on strategic governance and executive monitoring, while the execution board monitored deliverables and flow performance. The design of the flow stages, entry and exit criteria, Work in Progress (WIP) limits, service classes, and blocking criteria were defined.
Accompanying rituals were also established, with recurrent verification of blockages and flow health within the execution framework, in addition to weekly portfolio meetings for project entry, prioritization, pause, and completion decisions. The researcher acted as the system structuring agent, facilitator of rituals, and responsible for flow monitoring. To mitigate potential confirmation bias, source triangulation and temporal comparison were adopted.
The analytical procedures were structured in sequential steps. The As-Is scenario was diagnosed, followed by the design of the To-Be model and the institutionalization of system rules, with explicit definition of WIP limits, service classes, and blocking criteria. Finally, the periods before and after the intervention were compared. The quantitative analysis considered Kanban flow indicators (Lead Time, Throughput, time per stage, blockages, CFD) and executive governance indicators (completed projects, distribution by status).
The qualitative analysis was conducted based on content analysis, according to Bardin (1977), using records of participant observation, meeting minutes, and notes produced during follow-up rituals. The evidence was organized into categories related to project status clarity, prioritization capability, executive visibility, inter-area communication, and perception of roadblocks. Triangulation between documentary analysis, participant observation, and software records strengthened the consistency of the findings.
The ethical precepts were observed through the anonymization of organizational data and sensitive information related to participants and the company. Confidentiality was guaranteed in accordance with the principles applicable to the study. The research site was identified only by city and state, without details that could compromise the organization’s privacy.
3. Results and Discussion
The analysis of data extracted from the Businessmap software, used to operationalize the Kanban visual portfolio in the studied organization, revealed significant insights into the institutionalization of the method. The study compared two distinct periods: October to November 2025, representing the scenario before the intervention, and January to March 2026, which corresponds to the scenario after institutionalization. December was disregarded due to atypical seasonality that could distort comparative results. The findings were examined under two main dimensions: workflow predictability and executive visibility of strategic projects.
Pre-institutionalization scenario
The initial diagnosis, referring to the year 2025, showed that the organization managed between 15 and 25 strategic projects simultaneously, without an integrated view of dependencies or available productive capacity. Management was predominantly decentralized, based on spreadsheets, which resulted in a sharp fragmentation of information and low transparency regarding the actual progress of initiatives. This approach caused delays in technical stages that, in turn, provoked a cascading effect on other work fronts, culminating in rework and reactive prioritization.
The lack of visibility into roadblocks and production capacity made it difficult to anticipate risks, compromising the effectiveness of deliveries. Additionally, operational communication was costly, as project status needed to be reiterated in multiple forums and email exchanges. Qualitative reports indicated that senior leadership perceived a scenario of low effectiveness, with difficulty in promoting consistent progress on strategic initiatives. These findings corroborate Kerzner’s (2018) perspective, which points to information fragmentation as one of the main obstacles to project management maturity. The central problem, therefore, was not limited to delays, but to the absence of a structured flow management system, which compromised the predictability of deliveries and the capacity for strategic prioritization.
Process of institutionalization of Kanban
To mitigate the limitations identified in the previous scenario, a visual governance system based on the Kanban method was implemented, structured in two complementary levels. This structuring of the visual management system aligns with the principles of the Kanban method, which emphasizes the visualization of work and flow management as central mechanisms for continuous improvement, as proposed by Anderson (2010). The objective was to create an environment where work was transparent and flow was manageable.
At the portfolio level, the model was designed to support strategic governance, organizing initiatives by value nature and structuring a flow that encompassed everything from the strategic backlog to project completion. Each project was represented by a standardized card, containing essential information for decision-making, such as the responsible party, the strategic objective, and the next critical deliverable. This standardization aimed to enhance the clarity and consistency of information available to leadership.
At the execution level, the focus was directed to the flow of deliverables, with the explicit limitation of the quantity of simultaneously active projects and the clear definition of operational steps. Blockage management was formalized through visual records, indicating causes and responsible areas. Additionally, work in progress (WIP) limits, service classes, and periodic follow-up rituals were instituted. The application of WIP made the system’s capacity explicit, inducing more disciplined prioritization decisions, as proposed by Reinertsen (2009). The implementation of these practices promoted a structural change in management logic, shifting the focus from isolated projects to integrated workflow management.
Effects on the predictability of the flow
The analysis of monthly throughput by workflow revealed distinct patterns between tasks and initiatives. Task throughput remained relatively stable throughout the analyzed period, with variations consistent with seasonal effects. For example, in July 2025, 47 tasks were completed, while in March 2026, this number was 61. In contrast, initiative throughput showed significant growth in the post-institutionalization period, reaching its highest value in March 2026, with 31 initiatives completed, compared to 16 in July 2025. This behavior indicates an increase in completion capacity at the tactical level, suggesting greater efficiency in value delivery.
Time metrics reinforced this improvement trend. A reduction in the median cycle time was observed for both tasks and initiatives. For tasks, the median cycle time decreased from 4.0 days in the previous period (October-November 2025) to 3.1 days in the later period (January-March 2026). For initiatives, the reduction was even more significant, dropping from 12.8 days to 4.1 days. This decrease in the median indicates greater efficiency in the typical workflow after the institutionalization of Kanban, reflecting faster processing of demands.
Beyond the median, the analysis of the 85th percentile (P85) of cycle time for tasks demonstrated a significant reduction, from 18.7 days to 10.0 days. This decrease in P85 is an indicator of lower operational variability, suggesting that most tasks are being completed within a more consistent and predictable timeframe. However, for initiatives, the 85th percentile of cycle time increased from 31.0 days to 41.9 days, and the P85 lead time also grew from 34.5 days to 44.1 days. This increase in the upper percentiles for initiatives highlights the persistence of long-duration items, indicating that, although the average flow has improved, there are still challenges in managing more complex or blocked demands.
The distribution of times, analyzed through comparative boxplots, confirmed the reduction of the interquartile range for tasks and initiatives, which reinforces the greater consistency of the flow after the intervention. The median cycle time for tasks decreased from 4.0 days to 3.1 days, and for initiatives, from 12.8 days to 4.1 days. However, the presence of extreme values in the boxplots indicates that a portion of the demands still remains subject to blockages or greater complexity than the average. Together, these results point to a significant improvement in the central behavior of the system, with increased efficiency and reduced typical variability, but the persistence of long tails demonstrates that predictability is not yet fully stabilized, especially for strategic initiatives.
Effects on executive visibility
The implementation of the visual portfolio provided relevant advances in the executive visibility of projects. The structured visualization of the portfolio made it possible to make the value flow explicit, as proposed by Anderson (2010), facilitating the understanding of the initiatives’ progress. The distribution of cards by status allowed for immediate visualization of the quantity of prioritized, in progress, and completed initiatives, as well as identifying bottlenecks in the flow. For example, at a given moment, 4 prioritized initiatives, 2 in formalization, 7 in progress, and 28 finalized were observed, totaling 41 initiatives. For tasks, 12 prioritized, 6 in formalization, and 86 in progress were identified, summing up to 104 tasks. This type of visualization reduced the need for manual information consolidation and enabled a quick reading of the portfolio’s status.
The hierarchical structure between initiatives and tasks evidenced the relationship between the strategic and operational levels. Of the total 181 initiatives, 152 (84%) had linked tasks, which allowed for improved traceability between deliverables and strategic objectives. However, the presence of 29 initiatives (16%) without linked tasks and, more significantly, 317 tasks (55%) out of a total of 572 without a hierarchical link to initiatives, indicated the existence of reactive demands in the system. This portion of work not directly aligned with strategic objectives suggests that part of the workflow still operates in an unplanned manner, which can compromise the complete integration between strategy and execution.
From a qualitative point of view, an increase in the perception of clarity by the teams was observed, as they began to understand the progress of deliveries and their responsibilities more objectively. Structured visualization reduced the dependence on verbal communication and status meetings, optimizing the time and resources dedicated to alignment. Thus, visibility ceased to depend on individual interpretations and became based on shared visual artifacts, expanding the capacity for data-driven decision-making and promoting a more uniform understanding of project progress.
Limitations and partial effects
Despite the advances observed in flow predictability and executive visibility, relevant limitations were identified in the institutionalization process of Kanban. The persistence of initiatives with high duration, evidenced by the increase in the upper percentiles of time metrics for initiatives, indicates the need for deeper analysis of bottlenecks and management of complex items. This suggests that, although the average flow has improved, the most challenging exceptions and bottlenecks still require specific attention and more robust mitigation strategies.
Additionally, difficulties were observed in the full adoption of the tool by non-technical areas, highlighting the need for continuous training and more active system governance. Unequal adherence among teams, especially those with less familiarity with the logic of visual management, may compromise the overall effectiveness of the system. The dependence on operational discipline for data updating and maintaining information quality also proved to be a challenge, requiring an agent responsible for the continuous support of the system.
The existence of tasks not linked to strategic initiatives demonstrates that part of the workflow still operates reactively, which can compromise the complete integration between strategy and execution. This gap indicates that not all demands are being managed within the visual portfolio framework, resulting in work that, although necessary, does not directly contribute to strategic objectives explicitly. Thus, it is concluded that the model has been institutionalized with positive results in terms of predictability and visibility, but its maturity depends on the continuous evolution of usage practices, operational discipline, and team engagement.
In summary, the institutionalization of the Kanban method in the technology startup achieved the objective of proposing, applying, and analyzing a visual management system for monitoring multiple strategic projects. The results indicated a significant advance in flow predictability, especially due to the reduction in typical delivery variability and greater clarity on system capacity, although long-term initiatives still persist. In parallel, there was a substantial gain in executive visibility, with the centralization of information, standardization of cards, and clarification of progress stages, making decisions on prioritization, monitoring, pausing, and project completion more objective and data-driven. Kanban, therefore, was not limited to a visualization tool but was established as a coordination mechanism, clarifying flow rules, work-in-progress limits, blockages, and update responsibilities, which positively answers the research question about its contribution to executive predictability and visibility.
4. Conclusion
The study sought to propose, apply, and analyze the institutionalization of the Kanban method in the management of multiple strategic projects in a technology startup, with an emphasis on workflow predictability and executive visibility. It was verified that the implementation of a two-level visual governance system, with the explicit definition of workflow, the limitation of work in progress, and the adoption of monitoring rituals, contributed significantly to the organization of the workflow. An improvement in predictability was observed, evidenced by the reduction in the median cycle time for tasks and initiatives, and by the decrease in operational variability for tasks. Additionally, a substantial advancement in executive visibility was noted, through the centralization and standardization of portfolio information, which made decisions regarding prioritization, monitoring, and project completion more objective and data-driven. The practical contribution lies in the structuring of a coordination mechanism that transcends mere visualization, establishing clear workflow rules and update responsibilities, supporting strategic decision-making in a dynamic environment.
Despite the advances, relevant limitations were identified, such as the short analysis period after the system implementation and the dependence on operational discipline for maintaining data quality. The unequal adoption of the tool by non-technical areas and the persistence of long-duration initiatives were also noted, indicating the need for further depth in managing roadblocks and complex items. The existence of tasks not linked to strategic initiatives suggests that part of the workflow still operates reactively, compromising the full integration between strategy and execution. For future studies, it is recommended to extend the temporal analysis window to monitor the maturity of system use over a longer period, deepen the investigation of roadblocks, and examine with greater precision the relationship between executive visibility, workflow stability, and organizational performance.
Bibliographic References
ANDERSON, D. J. Kanban: Successful Evolutionary Change for Your Technology Business. Sequim: Blue Hole Press, 2010.
BARDIN, L. Análise de Conteúdo. Lisboa: Edições 70, 1977.
BERKUN, S. Making Things Happen: Mastering Project Management. Sebastopol: O’Reilly Media, 2008.
KERZNER, H. Gestão de Projetos: Uma Abordagem Sistêmica para Planejamento, Programação e Controle. 12. ed. São Paulo: Blucher, 2018.
REINERTSEN, D. G. The Principles of Product Development Flow: Second Generation Lean Product Development. Redondo Beach: Celeritas Publishing, 2009.
THIOLLENT, M. Metodologia da pesquisa-ação. 18. ed. São Paulo: Cortez, 2011.
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