The Data Leader's Playbook for Smarter Business Decisions
Agus Jamaludin, Deputy Data Science Manager, Petrosea
Agus Jamaludin, Deputy Data Science Manager, Petrosea
Agus Jamaludin is a passionate data leader who turns complex challenges into innovative, datadriven solutions. As Deputy Data Science Manager at Petrosea, he leads AI and analytics projects that improve operations and decision-making. With a hands-on, collaborative style, he brings clarity, innovation and impact to every stage of the data journey.
Bridging Strategic Goals with Data SolutionsAs Deputy Data Science Manager, I primarily bridge strategic business goals with data-driven solutions. I lead a multidisciplinary big data team comprising data analysts, data scientists, data engineers, BI developers and database administrators in developing advanced analytics, machine learning models, data integration, data visualization, data management and scalable data products that support decision-making across various departments.
My key responsibilities include:
1. Project Leadership: Overseeing end-to-end delivery of analytics projects from problem framing, data pipeline development, model deployment, to business integration.
2. Team Development: Mentoring and guiding junior data scientists to build technical and business acumen while fostering a collaborative and innovative team culture.
3. Stakeholder Engagement: Translating complex data insights into actionable strategies for senior leadership, ensuring alignment with business needs and priorities.
4. Data Strategy and Governance: Contributing to developing data architecture, data quality frameworks and long-term analytics strategy to ensure scalable and reliable data solutions.
My current focus areas are optimizing operations, implementing AI for real-time decision support systems and driving the adoption of data science practices across the organization.
Accelerating Insight with Clear VisualsData visualization, in my view, plays a critical role in the speed and quality of decision-making at managerial and executive levels. Well-designed visuals help leaders quickly grasp key insights, identify trends and detect anomalies, enabling faster and more informed decisions. However, a common challenge is that many users still manually create reports using tools like Excel. Without a strong foundation in visualization principles, the result is often cluttered visuals that lack clarity, especially problematic for executives who need concise, high-impact information.
Our big data team collaborates closely with the UI/UX team to design intuitive, user-centric dashboards and reports. This collaboration starts early in the project lifecycle, from planning through post-implementation to ensure visualizations are accurate, easy to navigate and tailored to user needs.
We follow a structured data lifecycle that begins with clearly defining the problem and understanding user pain points. After gathering requirements, we identify the most effective analytical approach. Not every issue requires advanced analytics or AI. A well-executed descriptive analysis, delivered through compelling visual storytelling, is often enough to support impactful decisions. Even the most advanced analytical or ML models are ineffective if poorly communicated insights. Effective data visualization bridges technical output with strategic action, making complex information accessible, relevant and valuable.
Balancing aesthetic design with functional clarity is essential when creating visualizations for decision-makers with varying levels of data literacy. My approach focuses on user-centered design to ensure visuals are purposeful, appealing and accessible to a broad audience.
The process starts with understanding the target users, their roles, data needs and how they engage with information. For senior executives, we prioritize clean dashboards highlighting KPIs, trends and alerts using intuitive visual elements. We offer drill-down views with filters and advanced visualizations for more technical users, such as analysts, often within the same reporting environment.
Once the audience is defined, we identify key metrics in close collaboration with stakeholders. We then work with our UI/UX team to apply design best practices while keeping functionality at the core. A unified design system ensures consistency in colors, fonts, icons and layout across all dashboards, reports and applications, creating a cohesive and professional experience.
Before deployment, we conduct user testing with stakeholders representing different levels of data literacy. Their feedback helps ensure the final product is polished, intuitive and practical, enabling fast, confident and data-driven decision-making across the organization.
Custom Web Applications for Complex Business NeedsWe use a combination of tools to deliver visualization-driven insights tailored to specific audiences and use cases. Our responsibilities extend beyond building dashboards; we also manage data integration to ensure that visualizations remain accurate and up to date.
The process starts with data acquisition using Azure Data Factory, which acts as middleware to extract data from multiple sources. This data is stored in our central warehouse through Azure Synapse, supporting efficient access and scalability.
Always prioritize simplicity and storytelling. This enhances the usability of dashboards and builds data confidence among non-technical users
For advanced analytics or complex preprocessing, we rely on Python. It enables customized data handling, exploratory analysis and internal visualizations for model development and technical communication. Power BI is our primary reporting tool, providing interactive dashboards for business users and executives. Its integration with Microsoft products, ability to handle large datasets and ease of use make it ideal for self-service analytics and operational monitoring.
When business needs exceed the scope of standard BI tools, such as requiring advanced user interaction or custom workflows, we build web-based applications with our software development team. These tools deliver highly customized visual experiences aligned with the needs of specific users or departments.
Communicate Insights with Clarity and ImpactI advise data science leaders to view data visualization as a strategic communication tool that can significantly influence decisions.
The process should begin with a clear understanding of the business problem, the supporting decisions and the story the data must tell. Involving stakeholders early, especially during the requirements gathering phase, ensures alignment with operational needs and data literacy levels.
Leaders should adopt a standardized design system with consistent color schemes, fonts, layout principles and familiar visualizations. Collaboration across teams, including data professionals, UI/UX designers and domain experts, is essential to create insightful and user-friendly outputs.
Equally essential is to invest in training across the organization. Many users still rely heavily on manual reporting tools like Excel and may lack the foundational knowledge to create or interpret compelling visualizations. Providing targeted training on choosing the right chart type for different kinds of data, for example, when to use a bar chart versus a line chart or present comparisons versus trends, can significantly improve the quality of insights shared across teams.
Always prioritize simplicity and storytelling. This enhances the usability of dashboards and builds data confidence among non-technical users.
Treat dashboards and reports as living products: gather feedback, observe usage patterns and continuously iterate to ensure they remain relevant, actionable and practical. Even the most sophisticated analytics are only as valuable as the clarity and impact of how their insights are communicated, and that’s where excellent visualization makes all the difference.
Weekly Brief
Read Also
