Challenge

Energy managers in large industries don't work with estimates. Decisions are based on financial data, consumption indicators, and default projections. Everything needs to be precise and, above all, available at the right time.

When the basis for these decisions is an Excel spreadsheet with dozens of tabs, chained formulas, and manual data entry, the risk ceases to be hypothetical.

Simple Energy operated precisely in this scenario. They had the data; the problem was the time required to transform it all into usable information and the constant reliance on human intervention throughout the process.

Simple Energy's operation is complex by nature. There are managers who run companies, which in turn interact with energy distributors. In this flow, financial, energy, and contractual data circulate constantly and need to be accurately consolidated to support decisions at every level.

All of this was done in Excel. Not in a simple way, but with a highly interdependent structure: tabs connected to each other, formulas that fed other formulas, and auxiliary data updated manually with each cycle. The complete processing took about two days. The problem wasn't just efficiency, it was risk.

Two days of processing meant accumulated delays and little margin for error. Any inconsistency along the way required starting over. There was a need to evolve, but there was no ready-made path. The market did not offer equivalent solutions for this type of operation, and there was no direct benchmark. This meant starting from scratch and, before any development, thoroughly understanding a specific technical vocabulary of the energy sector to translate the rules of spreadsheets into system logic.

Transformation_

The first step was to disassemble the spreadsheets, understanding the logic behind them. Each formula contained a business rule, each tab represented a dependency. Mapping this was essential to rebuilding the process in a structured way. From there, the Sim Connect calculation engine, the core of the system, was born.

This engine began to execute, in an automated and reliable way, everything that previously depended on formulas and manual data entry. The business rules were translated into SQL and organized into a dedicated service, isolated from the rest of the application.

To resolve the time bottleneck, the architecture was designed with task parallelism and multiple simultaneous connections to the database, allowing for more efficient processing of large volumes of information.

Data entry was also automated. Lambda functions in AWS were responsible for integrating and updating the database periodically, without manual intervention. On top of this, a .NET API began exposing the processed data for consumption.

In the frontend, developed in Angular, what was previously fragmented across spreadsheets was consolidated into a single system. Financial management, energy consumption per plant, and a consolidated view of contracts and projections now coexist in the same environment, organized into dashboards, tables, and filters.

During the project, a new feature emerged that did not exist in the previous model: version comparison. This allows the manager to select two power plants and a specific period to view the data side-by-side, both in terms of financial and energy consumption. This type of analysis was simply not possible before.

Benefits_

The most immediate impact was on processing time. What previously took about two days now takes approximately three hours. In practice, this means more analysis cycles, faster decisions, and less downtime waiting for data to be ready.

Automation eliminated the need for manual data entry, drastically reducing the risk of error throughout the entire calculation chain. Managers no longer work with manually manipulated data and now rely on a database generated by the system.

Furthermore, centralizing the information brought a new way of visualizing the operation. What was previously scattered across tabs was transformed into a consolidated, accessible, and configurable view. And the ability to compare plants in different periods added a layer of analysis that did not exist before.

Tools used_

• .NET
• AWS
• Angular

The .NET API centralizes access to processed data, Lambda functions ensure automatic information updates, integrating different sources continuously, and Angular provides the interface, bringing together dashboards, filters, and visualizations that transform complex data into actionable information.