Markus Ehrenmueller Jensen explains why data modeling, Microsoft Fabric, Power BI, SQL and user adoption still decide modern BI success today.Markus Ehrenmueller Jensen explains why data modeling, Microsoft Fabric, Power BI, SQL and user adoption still decide modern BI success today.
During a recent conversation on The Executive Outlook, Markus Ehrenmueller Jensen shared a reminder that many modern data teams need to hear.
The tools keep changing, but the fundamentals still decide whether business intelligence works.
Markus has spent decades working across the Microsoft data platform, long before Power BI, Azure and Microsoft Fabric became part of the modern data conversation. He has helped organizations across the full data pipeline, from finding and ingesting data to building ETL processes, data warehouses, analytics layers and reporting systems.
He is also the author of Data Modeling with Microsoft Power BI, which makes his perspective especially relevant at a time when many companies are rushing toward modern platforms but still struggling with slow dashboards, unclear models and limited business adoption.
His message is simple.
Before leaders blame the tool, they should look at the model, the structure and the way people use the data.
Markus did not begin his journey with cloud platforms or modern analytics tools. His curiosity started much earlier.
At the age of thirteen or fourteen, he was introduced to portable computers at school. These were not laptops in the way people understand them today. They were large machines with a detachable keyboard, a small monochrome display and floppy disks used to store programs and data.
The experience stayed with him.
He remembers writing programs in BASIC and seeing the computer respond to commands. If the code had a typo, it failed. If the command was right, the machine repeated the task exactly as instructed. That feeling of control and possibility made technology feel almost like science fiction in real life.
That early experience pulled him deeper into programming, SQL, business systems and eventually a long career in data. More than three decades later, he is still working with the same core idea that first attracted him to technology.
If you understand logic, the system can help people work better.
One of the most interesting parts of Markus’s story is how long SQL has remained relevant.
He has been writing SQL statements for more than thirty years. The platforms have changed. The tools have changed. The syntax has evolved in places. But the core idea remains powerful.
For Markus, SQL is still one of the most useful skills a data professional can learn because it gives people a common language to work with structured data. Even many newer database systems eventually introduce SQL style access because the language is widely understood and practical.
That matters for modern analytics teams.
A company may adopt Microsoft Fabric, lakehouses, warehouses, notebooks, Power BI, or other modern services. But if teams understand SQL, they already have a strong foundation for exploring, transforming and validating data across many environments.
In a market that keeps chasing new tools, Markus’s point is refreshing. Some skills stay valuable because they solve a problem that has not gone away.
People still need to ask questions from data.
Markus describes Microsoft Fabric as both familiar and genuinely new.
On one side, many parts of Fabric are built on ideas and services that Microsoft users may already recognize. If someone has worked with Azure SQL, data flows, Power BI, pipelines, or related services, they will see familiar patterns inside Fabric.
But on the other side, Fabric brings those pieces together under one umbrella. That matters because many data projects become complex not because one tool is difficult, but because teams are forced to manage many tools, interfaces, licenses, security settings and integration points at the same time.
Fabric tries to reduce that complexity.
A major part of this is OneLake. Markus explains that different workloads inside Fabric can read from and write to the same shared data foundation. A Python user can work with notebooks. A SQL user can query tables. A Power BI user can build reporting models. Each person can use the skills they already have while working around the same data environment.
That is powerful because it reduces the friction between teams.
Instead of constantly copying data between systems or building one connector after another, Fabric makes it easier for different roles to collaborate around the same data. For organizations with mixed skill sets, that can be a major shift.
The value of OneLake is not only technical. It is also organizational.
In many companies, data work gets slowed down because different people work in different tools and create separate versions of the same thing. One team prepares data in Python. Another works in SQL. Another build reports in Power BI. Each handoff creates risk, delay and confusion.
Markus sees Fabric as a way to bring those workflows closer together.
If a data engineer prepares a table in a lakehouse, another person with SQL skills can query it as if it were a database table. A Power BI user can then bring that same table into a semantic model and create reports from it. The same foundation supports different ways of working.
This is where Microsoft Fabric becomes more than another data platform. It becomes a way to let teams collaborate without forcing everyone to use the same tool or skill set.
For business leaders, that distinction matters. The goal is not only to modernize technology. The goal is to make data work easier across teams.
Markus also gives a practical view of Microsoft Fabric cost.
He does not present Fabric as automatically cheaper or more expensive than every other option. Instead, he explains that Fabric works through capacity and consumption. Teams pay capacity, and workloads consume units based on how they are used.
That makes cost estimation difficult without testing real use cases.
His practical advice is to start with a Fabric trial, build proof of concepts, observe consumption patterns and learn how different workloads behave. A transformation done in SQL may consume differently from the same transformation done in Spark. The only way to understand that properly is to test, measure and tune.
He also points to the importance of monitoring Fabric capacity through the metrics app. Leaders and teams need visibility into which workloads are consuming capacity, when usage is rising and where performance tuning may reduce cost.
This is an important executive lesson. Modern platforms do not remove the need for cost discipline. They make cost visibility and active management even more important.
Markus also gives a balanced view of cloud versus premises infrastructure.
If an organization has very stable and predictable workloads, premises systems may sometimes be cheaper because the hardware is already owned and usage is consistent. But if workloads rise and fall, clouds can be more useful because services can be scaled up and down as needed.
For Markus, the bigger difference between Cloud v/s On Premises is not only cost. It is speed and flexibility.
In the cloud, a team can create a lakehouse or test a new service quickly. On premises, that same experiment can require hardware, installation, configuration, access, maintenance and time. The cost of waiting can be just as important as the cost of infrastructure.
He also makes a thoughtful point about talent. Many younger professionals entering the workforce are cloud natives. They are used to clicking, testing, learning, and shutting down services quickly. If a company forces them into slow infrastructure processes, it may struggle to attract or retain the kind of talent modern data work requires.
That makes cloud a capability decision, not only a pricing decision.
The strongest lesson from Markus’s conversation is about data modeling.
He believes data modeling is still underestimated, especially in Power BI and now in the Fabric era. Too many teams get access to data and immediately start building visuals. They connect tables, create reports, add filters and only later realize that the dashboard is slow, the calculations are difficult and the logic is fragile.
Markus has lived this himself.
He described his first Power BI report that went into production. He loaded data directly from the system, added many bidirectional filters because that was what the customer wanted and expected everything to perform well because Power BI works in memory. Instead, the report became slow and difficult to manage.
That experience taught him an important lesson.
Do not rush into visualization before understanding the data.
In Power BI, a strong model often means clear facts, dimensions, relationships and structure. When the model is right, analytics becomes faster, DAX becomes easier and users have a better experience. When the model is wrong, even simple calculations can become unnecessarily hard.
This is why data modeling still decides BI success.
Markus also talks about what he calls the last mile in business intelligence.
An organization may have strong hardware, a centralized data warehouse and clean technical architecture. But people may still struggle to get useful information out of it. That final step between having data and using data is where many BI initiatives succeed or fail.
Sometimes the issue is the data model. Sometimes it is a reporting tool. Sometimes it is education. Sometimes users try to use a new tool with the mindset of the old one.
That is a change management problem as much as a technology problem.
Markus explains that business intelligence often involves two groups that do not naturally speak the same language. IT teams understand systems, structure, governance and architecture. Domain experts understand business questions, context, decisions and daily pressure. If these groups do not work together, the organization will struggle to turn data into action.
This is why the last mile matters so much.
BI is not successful when a report is published. It is successful when people can use it to make better decisions.
Markus also brings an important perspective on enterprise BI and self-service BI.
Some reporting needs should be centralized. For example, structured reports such as official finance reports, balance sheets and regularly used operational lists should be governed and consistent. These are not reports that every department should rebuild separately.
But business changes quickly. Unexpected events happen. New questions arise. Teams need the ability to explore, test and analyze data without waiting weeks for every answer.
That is where self-service BI becomes important.
The mistake is treating enterprise BI and self-service BI as enemies. Organizations need both. Enterprise BI gives consistency. Self-service BI gives agility. The real work is finding the right balance and making sure business users have access to the right models, the right data and the right education.
Power BI, in Markus’s view, sits in the middle of that balance. It supports enterprise reporting and self-service analysis, but the organization still has to decide how both work together.
Across the full conversation, Markus Ehrenmueller Jensen is not simply talking about Microsoft Fabric, SQL, or Power BI as tools. He is talking about the discipline behind modern analytics.
His experience shows that platforms change, but the important questions remain familiar.
Is the data modeled properly?
Can teams work from a shared foundation?
Do users understand how to use the tools?
Are IT and business experts working together?
Is the platform designed for real decisions, not just technical completion?
For leaders investing in Microsoft Fabric, Power BI, or the broader Microsoft data platform, Markus’s message is practical and important. Do not assume a modern tool automatically creates modern analytics. The value comes from how well the data is structured, how clearly it is modeled, how easily teams can collaborate, and how confidently users can reach the information they need.
The future of business intelligence may be powered by new platforms. But its success still depends on the fundamentals.
Understand the data. Build the model. Educate the users. Then build the dashboard.
Want to hear more conversations with leaders shaping data, analytics, Microsoft Fabric and business intelligence? Explore more on The Executive Outlook.
Listen in for business strategies and interviews with business experts to help you build a business and a life that you love.
About Us
