Reflections on MIS 587 - Business Intelligence
Introduction to Business Intelligence
Business Intelligence is not just a collection of tools and processes. It is a unified and value driven approach that strengthens decision making at every level of an organization. As a Business Intelligence developer, throughout this MIS 587 course, I gained a deeper appreciation for how complex and interconnected the Business Intelligence ecosystem truly is. Data warehousing, modeling, ETL/ELT processes, dashboards, web analytics and insights, and network analysis all contribute to a broader goal of transforming raw information into meaningful knowledge. This course encouraged me to look beyond definitions and reflect on how these concepts apply directly to real organizational challenges.
Data Warehouse Foundation
My experience with data warehousing was one of the areas where the concepts in class closely aligned with my real work responsibilities. A warehouse is much more than a storage system. It is a structured and purpose-driven environment where data must be organized, cleaned and modeled to support analytics. In this class, we explored different warehouse architectures and their role in supporting an organization’s strategy for Business Intelligence. Whether the system follows a traditional data warehouse, a data lake, or a modern Lakehouse approach, the objective remains the same. Data must be accessible, trustworthy and modeled correctly.
I also realized how strongly warehousing decisions connect to business needs. Finance relies on consistent structures for reporting. Marketing depends on historical trends. Operations need detailed and timely data. A warehouse becomes the foundation for all these requirements, and this course showed how much thoughtful planning goes into building that foundation.
Dimensional Modeling
Dimensional modeling was one of the most valuable topics for me. Since I regularly work with Oracle Data Warehouse and Informatica PowerCenter, learning the theory behind fact tables, dimension tables, star schemas, and snowflake structures helped reinforce what I see in practice. A well-designed data model simplifies analytics and improves the experience for business users, especially those working in a self-service environment or workspace. I gained a deeper appreciation for how modeling decisions influence every layer of analytics. Granularity determines how detailed reports can be. Dimension design affects how flexible analysis becomes. Even something as simple as how dates are structured can change the user experience. When a warehouse is built with clear relationships and appropriate levels of detail, users can explore data confidently without depending on developers for every request. This reduces bottlenecks and increases data literacy across the organization. At the same time, it introduces a challenge because BI developers must understand the data flow and the underlying processes to build an effective dimensional model.
Modern ETL/ELT and Real Time Data Analytics
The course also emphasized the importance of ETL and ELT processes. Traditional ETL systems rely on scheduled refreshes that run daily or hourly. This has been the standard for many years, but modern organizations now expect data that is updated in near real time. Sales teams want dashboards to reflect performance within minutes. Operations teams need immediate visibility into inventory and transactions. These expectations highlight the limitations of older ETL systems and data integration.
Real time replication technologies, such as Oracle GoldenGate and Qlik Replicate have become essential because they continuously stream data from source to destination systems. This reduces the delay that traditional batch ETL creates and enables organizations to move closer to real time analytics.
However, these tools introduce another layer of complexity. Real time replication transfers data exactly as it exists in the source, which shifts modeling and business logic to downstream processes. Some organizations try to use replication tools as real time ETL by loading data directly into modeled tables, but this creates significant architectural challenges. Real time transformations require constant processing power, create risks of system delays and can produce performance and database issues. The more logic that is embedded into a real time pipeline, the more difficult it becomes to ensure reliability and consistency. This reminded me that even as technology improves, strong data modeling, governance and architecture are still essential. Real time data movement cannot replace solid design principles.
Analytics and Visualization Tools
Working with Business Intelligence tools in this course strengthened my understanding of how analytics are delivered to users. I mainly use Power BI and SAP BusinessObjects at work, but experimenting with Tableau offered a fresh perspective. Tableau’s method of creating separate workbooks and combining them into complete dashboards helped organize the development process in a clear and efficient way.
One important lesson from this course is that visualization tools are not just reporting platforms. They are storytelling platforms. The layout, color choices, visual types and structure can completely change how a user interprets the information. A well-designed dashboard guides attention and reveals insights. A poorly designed one creates confusion. Regardless of the platform, visual analytics plays an essential role in helping users explore data, monitor performance indicators, identify trends and find insights that tables alone cannot provide.
Web Analytics and Digital Insight
Web analytics expanded my understanding of Business Intelligence even further. Analyzing website traffic showed how global and diverse online engagement can be. Even a small website, such as the Google Merch store receives visitors from all over the world. Through geographic data, user flow, device usage and traffic sources, web analytics offers valuable insight into digital behavior.
What surprised me is how small website adjustments can produce measurable changes in user activity. Businesses use these insights to optimize marketing campaigns, enhance user experience, adjust product strategies and increase conversions. With the rapid growth of online platforms and social media, the ability to interpret web behavior has become a major skill for organizations of all industries.
Network Analysis and Gephi
One of the most unique and surprising parts of the course was working with Gephi for network analysis. When I first loaded the dataset, the graph appeared overwhelmingly dense, filled with many nodes and edges. It looked impossible to interpret. Applying modularity classes, degree centrality and betweenness centrality transformed the visualization and made patterns clear. Clusters formed, communities became visible and key connections emerged.
Network analysis revealed relationships that would never appear in a table. This approach has strong applications in social media analytics, fraud detection, recommendation systems, supply chain patterns and organizational communication. It showed me that Business Intelligence does not always revolve around tables and dashboards. Sometimes the structure of relationships reveals more than the values themselves.
Different Responsibilities for BI Developers
The Future of BI and Cloud Architecture
Reflecting on the future of Business Intelligence, it seems clear that the field is moving strongly toward cloud-based architecture, such as Microsoft Fabric and Azure Synapse, Snowflake and Databricks. These platforms allow organizations to combine structured and unstructured data in the same environment, which aligns with the growing popularity of the Lakehouse model.
Cloud platforms simplify infrastructure, improve scalability and support both machine learning and traditional analytics. However, cloud adoption also brings challenges. Security concerns, export compliance requirements and rising costs remain major obstacles. Some organizations choose hybrid environments where sensitive data stays on premises while cloud resources are used for scalable analytics.
Even with these concerns, the direction of the industry is clear. Organizations want faster insights, more automation and more flexibility. Business Intelligence professionals must develop skills in cloud architecture, real time data pipelines, modeling, governance and the integration of artificial intelligence. The specific tools may change in the future, but the principles of data quality, accessibility and reliability will always remain the foundation of the field.
Conclusion
Throughout the semester, I realized how deeply Business Intelligence influences organizational strategy. Data silos slow decision making and create inconsistencies. Poor data quality causes mistrust. Slow refresh cycles reduce operational visibility. Weak modeling structures create confusion and inefficiencies. The concepts we studied in MIS 587 showed how Business Intelligence addresses these challenges in a structured and scalable way.
This course strengthened my understanding of both the technical and conceptual sides of Business Intelligence. It helped me connect the material to real world experience and encouraged me to think carefully about the future of the field. Business Intelligence continues to grow in importance as artificial intelligence and cloud technologies become more integrated into daily operations. For me, MIS 587 confirmed that working with data is not only part of my current job but also a long-term career direction. The knowledge and skills I gained in this course will support my development as I continue exploring data engineering, analytics and Business Intelligence architecture.
References
Schneider, J., Gröger, C., Lutsch, A., Schwarz, H., & Mitschang, B. (2024). The Lakehouse: State of the Art on Concepts and Technologies. Springer Nature Link. https://link.springer.com/article/10.1007/s42979-024-02737-0
Business Intelligence is not just a collection of tools and processes. It is a unified and value driven approach that strengthens decision making at every level of an organization. As a Business Intelligence developer, throughout this MIS 587 course, I gained a deeper appreciation for how complex and interconnected the Business Intelligence ecosystem truly is. Data warehousing, modeling, ETL/ELT processes, dashboards, web analytics and insights, and network analysis all contribute to a broader goal of transforming raw information into meaningful knowledge. This course encouraged me to look beyond definitions and reflect on how these concepts apply directly to real organizational challenges.
Data Warehouse Foundation
My experience with data warehousing was one of the areas where the concepts in class closely aligned with my real work responsibilities. A warehouse is much more than a storage system. It is a structured and purpose-driven environment where data must be organized, cleaned and modeled to support analytics. In this class, we explored different warehouse architectures and their role in supporting an organization’s strategy for Business Intelligence. Whether the system follows a traditional data warehouse, a data lake, or a modern Lakehouse approach, the objective remains the same. Data must be accessible, trustworthy and modeled correctly.
I also realized how strongly warehousing decisions connect to business needs. Finance relies on consistent structures for reporting. Marketing depends on historical trends. Operations need detailed and timely data. A warehouse becomes the foundation for all these requirements, and this course showed how much thoughtful planning goes into building that foundation.
Dimensional Modeling
Dimensional modeling was one of the most valuable topics for me. Since I regularly work with Oracle Data Warehouse and Informatica PowerCenter, learning the theory behind fact tables, dimension tables, star schemas, and snowflake structures helped reinforce what I see in practice. A well-designed data model simplifies analytics and improves the experience for business users, especially those working in a self-service environment or workspace. I gained a deeper appreciation for how modeling decisions influence every layer of analytics. Granularity determines how detailed reports can be. Dimension design affects how flexible analysis becomes. Even something as simple as how dates are structured can change the user experience. When a warehouse is built with clear relationships and appropriate levels of detail, users can explore data confidently without depending on developers for every request. This reduces bottlenecks and increases data literacy across the organization. At the same time, it introduces a challenge because BI developers must understand the data flow and the underlying processes to build an effective dimensional model.
Modern ETL/ELT and Real Time Data Analytics
The course also emphasized the importance of ETL and ELT processes. Traditional ETL systems rely on scheduled refreshes that run daily or hourly. This has been the standard for many years, but modern organizations now expect data that is updated in near real time. Sales teams want dashboards to reflect performance within minutes. Operations teams need immediate visibility into inventory and transactions. These expectations highlight the limitations of older ETL systems and data integration.
Real time replication technologies, such as Oracle GoldenGate and Qlik Replicate have become essential because they continuously stream data from source to destination systems. This reduces the delay that traditional batch ETL creates and enables organizations to move closer to real time analytics.
However, these tools introduce another layer of complexity. Real time replication transfers data exactly as it exists in the source, which shifts modeling and business logic to downstream processes. Some organizations try to use replication tools as real time ETL by loading data directly into modeled tables, but this creates significant architectural challenges. Real time transformations require constant processing power, create risks of system delays and can produce performance and database issues. The more logic that is embedded into a real time pipeline, the more difficult it becomes to ensure reliability and consistency. This reminded me that even as technology improves, strong data modeling, governance and architecture are still essential. Real time data movement cannot replace solid design principles.
Analytics and Visualization Tools
Working with Business Intelligence tools in this course strengthened my understanding of how analytics are delivered to users. I mainly use Power BI and SAP BusinessObjects at work, but experimenting with Tableau offered a fresh perspective. Tableau’s method of creating separate workbooks and combining them into complete dashboards helped organize the development process in a clear and efficient way.
One important lesson from this course is that visualization tools are not just reporting platforms. They are storytelling platforms. The layout, color choices, visual types and structure can completely change how a user interprets the information. A well-designed dashboard guides attention and reveals insights. A poorly designed one creates confusion. Regardless of the platform, visual analytics plays an essential role in helping users explore data, monitor performance indicators, identify trends and find insights that tables alone cannot provide.
Web Analytics and Digital Insight
Web analytics expanded my understanding of Business Intelligence even further. Analyzing website traffic showed how global and diverse online engagement can be. Even a small website, such as the Google Merch store receives visitors from all over the world. Through geographic data, user flow, device usage and traffic sources, web analytics offers valuable insight into digital behavior.
What surprised me is how small website adjustments can produce measurable changes in user activity. Businesses use these insights to optimize marketing campaigns, enhance user experience, adjust product strategies and increase conversions. With the rapid growth of online platforms and social media, the ability to interpret web behavior has become a major skill for organizations of all industries.
Network Analysis and Gephi
One of the most unique and surprising parts of the course was working with Gephi for network analysis. When I first loaded the dataset, the graph appeared overwhelmingly dense, filled with many nodes and edges. It looked impossible to interpret. Applying modularity classes, degree centrality and betweenness centrality transformed the visualization and made patterns clear. Clusters formed, communities became visible and key connections emerged.
Network analysis revealed relationships that would never appear in a table. This approach has strong applications in social media analytics, fraud detection, recommendation systems, supply chain patterns and organizational communication. It showed me that Business Intelligence does not always revolve around tables and dashboards. Sometimes the structure of relationships reveals more than the values themselves.
Different Responsibilities for BI Developers
Business Intelligence developer responsibilities vary widely across organizations. Some companies focus on data analysis and rely on developers to build reports and dashboards for business users. Other companies place more emphasis on backend processes, where developers focus on moving data from source systems to destination systems and preparing it for analytics.
There are also organizations where a Business Intelligence developer manages the entire life cycle. This includes extracting and loading data, designing the data model in the warehouse, building database objects, creating semantic layers, and finally developing reports or delivering data to end users. This full-life-cycle approach requires a combination of technical expertise, business understanding, and strong communication skills. I personally focus on both the backend and frontend aspects of the BI role. I find that working across the full spectrum of BI allows me to advance my skill set and gain experience in all areas. It not only improves my technical skills but also strengthens my abilities in project management, communication, and business operations. From managing timelines and due dates to staying on budget, gathering requirements, coordinating with business users, and ultimately building the final product, I have learned to integrate technical work with business priorities effectively.
The Future of BI and Cloud Architecture
Reflecting on the future of Business Intelligence, it seems clear that the field is moving strongly toward cloud-based architecture, such as Microsoft Fabric and Azure Synapse, Snowflake and Databricks. These platforms allow organizations to combine structured and unstructured data in the same environment, which aligns with the growing popularity of the Lakehouse model.
Cloud platforms simplify infrastructure, improve scalability and support both machine learning and traditional analytics. However, cloud adoption also brings challenges. Security concerns, export compliance requirements and rising costs remain major obstacles. Some organizations choose hybrid environments where sensitive data stays on premises while cloud resources are used for scalable analytics.
Even with these concerns, the direction of the industry is clear. Organizations want faster insights, more automation and more flexibility. Business Intelligence professionals must develop skills in cloud architecture, real time data pipelines, modeling, governance and the integration of artificial intelligence. The specific tools may change in the future, but the principles of data quality, accessibility and reliability will always remain the foundation of the field.
Conclusion
Throughout the semester, I realized how deeply Business Intelligence influences organizational strategy. Data silos slow decision making and create inconsistencies. Poor data quality causes mistrust. Slow refresh cycles reduce operational visibility. Weak modeling structures create confusion and inefficiencies. The concepts we studied in MIS 587 showed how Business Intelligence addresses these challenges in a structured and scalable way.
This course strengthened my understanding of both the technical and conceptual sides of Business Intelligence. It helped me connect the material to real world experience and encouraged me to think carefully about the future of the field. Business Intelligence continues to grow in importance as artificial intelligence and cloud technologies become more integrated into daily operations. For me, MIS 587 confirmed that working with data is not only part of my current job but also a long-term career direction. The knowledge and skills I gained in this course will support my development as I continue exploring data engineering, analytics and Business Intelligence architecture.
References
Schneider, J., Gröger, C., Lutsch, A., Schwarz, H., & Mitschang, B. (2024). The Lakehouse: State of the Art on Concepts and Technologies. Springer Nature Link. https://link.springer.com/article/10.1007/s42979-024-02737-0
Tan, A. (2025, July 21). How AI is shaping the future of business intelligence. Computer Weekly. https://www.computerweekly.com/feature/How-AI-is-shaping-the-future-of-business-intelligence
Sudha Ram, Class Notes, MIS 587, Introduction to Business Intelligence, slide 7-26
Sudha Ram, Class Notes, MIS 587, Data Warehouse Design Cycle, slide 3-13
Sudha Ram, Class Notes, MIS 587, Advanced Star Schema Design, slide 4-20
Sudha Ram, Class Notes, MIS 587, Introduction to Web Analytics, slide 4-27
Sudha Ram, Class Notes, MIS 587, Introduction to Networks,
slide 4-13
Jim, K. (2025, September 4). What is the role of data warehouse in Business Intelligence?Airbyte. https://airbyte.com/data-engineering-resources/business-intelligence-data-warehouse
Onome, D. (2025, May 16). Real-Time ETLT: Meet the Demands of Modern Data Processing. DEV Community. https://dev.to/nomzykush/real-time-etlt-meet-the-demands-of-modern-data-processing-bke
Hi Faris!
ReplyDeleteYour reflection on the different areas of Business Intelligence shows how much depth there is behind the tools and processes we use every day. I really liked how you connected each concept back to your own experiences. It’s clear that the course helped reinforce the technical foundation you already had while also pushing you to think about BI from a broader perspective. Reading about how you balance backend work, data modeling, semantic layers, and visualization made me appreciate how multi-layered the BI role truly is.
With all of these topics covered, which topic ended up being your favorite, and which one do you see yourself applying the most outside of class as you continue growing in your BI career?
Good luck with the rest of your program and in pursuing your degree!
Hi Faris,
ReplyDeleteThank you for sharing your post. This is an exceptionally thorough reflection that demonstrates a deep understanding of how the core components of Business Intelligence connect across architecture, modeling, analytics, and organizational strategy. I especially liked how you linked dimensional modeling, ETL/ELT pipelines, and cloud technologies to the real challenges BI developers face today. Your examples made the theory very practical. Your discussion of network analysis and the future of BI shows strong insight into how the field is evolving toward real-time data, Lakehouse architectures, and AI-driven analytics. Overall, this is a clear, well-structured, and thoughtful reflection that shows strong mastery of the MIS 587 materials and how they apply directly to your current and future BI work.
Iman Arian
Your reflection does a great job highlighting how interconnected the different components of Business Intelligence really are. I appreciated how you tied each concept data warehousing, dimensional modeling, ETL/ELT, visualization tools, web analytics, and network analysis—back to your real-world experience. It really reinforces how BI is not just a technical discipline but a strategic one.
ReplyDeleteI especially liked your discussion about real-time data pipelines and the challenges that come with relying too heavily on replication tools for transformation. Your point about the importance of strong modeling and governance, even in modern architectures, is something many organizations overlook as they rush toward real-time analytics.
Overall, this was a thoughtful and comprehensive reflection. It’s clear that MIS 587 helped solidify the foundation you already had while expanding your perspective on how BI fits into organizational strategy.