Case · Data & Governance

Integrated Budget Governance

From fragmented SAP and GDTec extraction to a single view for monitoring epics, periods, balances and budget execution with far less manual work.

SAPGDTecPower QueryExcel
Real implementationProcess applied in a large enterprise environment
How the solution was built

From source updates to comparison between actual and simulated scenarios.

The flow below summarizes the case architecture: what still requires manual intervention, what is automated, and how the solution separates official data from planning scenarios.

Simplified view of the real process
01 · Source systems
Manual step

Current official baseline

Actual values still need to be updated manually in the corporate system before extraction.

CSV
CSV exportofficial source for the current scenario
Supplementary

Additional data

Reflects information from the primary system and adds data not directly available from the same source.

JSON PDF Excel
02 · Controlled input

Shared folder

Extracted files are organized in a shared network folder and become a recurring input to the solution.

📁 budget_base ├─ SAP / CSV └─ GDTec / JSON · PDF · XLSX
03 · Automated processing
P

Power Query

Performs cleaning, consolidation, source matching and preparation of current and simulated views.

1 read files 2 clean and type 3 rules and composition 4 refresh views
04 · Two views of the same budget
X
OFFICIAL

CURRENT SAP

Consolidated view of the actual scenario recorded in the official sources.

AuthorizedR$ 1,52 mi
ActualR$ 782 mil
BalanceR$ 356 mil
YearMonthEpicContract
SIMULATION

SIMULATED SAP

Separate area for testing assumptions without changing the official view.

New contract Jul → Nov R$ 80k / month
Refreshepic recalculated
Observed operational impact From 3–5 hours to 25–30 minutes

For more complex analyses, manual work could reach approximately 8 hours.

BEFORE 3–5 h typical manual preparation

Contract-by-contract consultation, consolidation of multiple epics, comparison of planned, actual and authorized values, and creation of new Excel views.

Up to ~8 h in higher-complexity scenarios
automation + reusable structure
AFTER 25–30 min complete solution refresh

Source updates, file extraction, base loading and refresh of the entire processing and visualization structure.

Available 24/7 for consultation and simulation
The main gain was not only reducing time.

The solution eliminated the need to rebuild the analysis for every new question: the structure is already prepared to answer different slices and scenarios.

*Estimate based on the pre-automation operational flow. The 25–30 minute duration corresponds to the current observed process.
Official datasource extraction
Processingautomated consolidation
Simulated scenarioassumptions separated from actuals
The value of the solution is not only consolidating data, but turning new questions into queries — not new manual work.
The problem

The information existed. The management view did not.

Budget data was available across corporate platforms, but turning it into a useful management view required extensive querying, copying, consolidation and manual calculations.

To monitor budgets by epic, contract, month or year, the process could require checking items one by one, moving values into spreadsheets and rebuilding the consolidated view. The broader the period, the greater the operational effort and risk of inconsistencies.

Before the solution

×Fragmented consultation across different sources and periods.
×Manual consolidation of epics, contracts and values.
×Low flexibility to switch between month, year, epic or full view.
×Time spent preparing information before analysis could even begin.
The reasoning behind the architecture

Each layer solves a different problem.

The technical design separates extraction, storage, processing and consumption so they do not depend on the same manual step.

01

Separate source from analysis

SAP and GDTec remain the official sources. The solution does not change those systems; it organizes how their data is consumed.

02

Create a recurring input

The shared folder becomes a predictable entry point. New files can feed the process without rebuilding the structure from scratch.

03

Centralize processing

Power Query centralizes cleaning, typing, rules and consolidation. Complexity stays in the technical layer — not with the person consuming the information.

04

Deliver a simple view

Decision-makers work with two clear views: the current official scenario and a simulation area to test assumptions without changing actuals.

What the solution enables

Less preparation. More analysis.

01

Consolidated view

Analyze a specific epic or the full portfolio without manually rebuilding the base for each analysis.

02

Time filtering

Switch between months and years while keeping history available for analysis.

03

Budget view

Monitor authorized, actual, planned values and balances in a single view.

04

Reusable base

New exports enter the flow without requiring manual reconstruction of the consolidated structure.

Impact

Information stops being an assembly task.

The main gain is not “having a better spreadsheet”.

It turns multiple operational sources into an information layer ready for consultation, comparison and decision-making — reducing dependence on recurring manual manipulation.

Less manual consolidation
Greater consistency across periods
History accessible in a single view
Faster, more flexible management analysis
Technical backstage

The decisions behind the solution.

Some project decisions show where the real complexity was — without turning the case into implementation documentation.

Folder-oriented architecture

Exported files become a recurring input layer. Power Query reads, processes and consolidates the content without requiring each file to be handled individually.

Separation between source and view

The processing layer keeps source complexity away from users who only need to consult budgets, history and balances.

SAPGDTecPower QueryExcelData consolidationGovernance
How I work

Different projects. The same principle:

understand the context · structure the problem · build a reliable solution