My boss had recently asked me to start working on the Annual Operating Plan (AOP). This is one of the most important deliverables in FP&A, the financial blueprint for the upcoming fiscal year.

It was my first time building it, and the level of detail was massive. I needed to model the full income statement, with thousands of variables to consider: clients, products, months, accounts, and more.

I was familiar with rigid financial models made up of multiple rows and columns, with references pointing to different schedules and sometimes different sheets. These models work well when data sources are limited and inputs are specific. But they begin to break down when complexity increases.

When I started working on the AOP, I was overwhelmed by the amount of information I had to process. Managing the AOP without Power Query felt unsustainable. Starting with the number of tabs and tables in my Excel file, it became nearly impossible to audit all references. Fixing broken or inconsistent links in a model that is spread in 20 tabs is something I don’t wish anyone.

I knew I had to find a way to process the data more elegantly and without errors.

In my search for better workflows, I stumbled upon Power Query. At first, it felt like learning a new language. In a way, it was. Power Query uses the M language and understanding how it transforms data required a different way of thinking.

It took some time to get comfortable with the interface and, more importantly, to understand how to structure data so the data model could process it efficiently. I had to stop thinking in terms of spreadsheets and start thinking in terms of structured datasets.

One of the most challenging concepts to grasp was the Unpivot transformation. Unpivot converts wide, matrix-style tables—where values are spread across multiple columns—into a structured, tabular format where each row represents a single observation. This long format is what data models typically require in order to aggregate and analyze information properly.

Once I understood this, everything changed.

With this new knowledge, the AOP was no longer chaotic. My reports became significantly more powerful and scalable. I was able to:

• Append multiple tables into a single fact table
• Expand dimension tables to segment data across multiple attributes
• Create customized measures in DAX within the data model
• Build financial models with thousands of inputs that process large volumes of data consistently

Instead of managing complexity manually, I was designing systems that handled complexity for me.

Ultimately, Power Query changed my game in FP&A. I became faster, more precise, and most importantly, I reduced modelling risks.

Without this knowledge, I would not be able to build the financial models I rely on today.

If there is one recommendation I would give to any aspiring finance professional, especially those working with large datasets, it would be to learn Power Query. It is not just a productivity tool. It is a shift in how you approach data and financial modeling.