EuroFoods S.A. — FP&A Performance Dashboard

An agency-grade Streamlit application that replicates and extends the Power BI dashboard from the Session 5 case-study brief. It analyses two years of EuroFoods sales (Jan 2023 – Dec 2024) across 3 countries, 4 categories, 4 channels and 20 products, and projects Q1 2025.

Quick start

# 1. Install dependencies
pip install -r requirements.txt

# 2. Make sure the workbook sits next to app.py
#    (case_study_eurofoods.xlsx — included in this repo)

# 3. Launch
streamlit run app.py

The app opens at http://localhost:8501.

What's inside

A 4-page dashboard with global slicers (Country / Year / Quarter / Category / Segment) in the sidebar:

Page	Highlights
Executive Summary	KPI cards (Revenue, Gross Margin, Margin %, YoY); Revenue by country actual vs budget; monthly revenue overlay 2023 vs 2024; segment donut.
Category Deep-Dive	Revenue category × country (stacked); 24-month margin-% lines with target reference; margin matrix with conditional formatting; Top-10 products; products below target margin.
Budget Variance	Budget → Actual waterfall by category; quarterly variance by country; Country × Category matrix (Actual / Budget / Variance € / Var %); price-vs-volume bridge.
Forecast & What-If	Growth slider (−10% → +20%); Jan–Mar 2025 projection (central / optimistic / pessimistic); per-segment scenario simulator; back-solve for the €2.5M 2025 target.

Project structure

.
├── app.py                  # Streamlit app: routing, pages, charts
├── src/
│   ├── data.py             # Star schema: load workbook, monthly aggregation, variance bridge
│   ├── metrics.py          # Measures: KPIs, YoY, variance, forecast, scenarios
│   └── theme.py            # Palette, Plotly template, number formatting, KPI-card CSS
├── case_study_eurofoods.xlsx
├── data_profile.md         # Step 0 — data profiling report (types, keys, anomalies)
├── findings.md             # Step 4 — written answers to the 11 analytical questions
├── requirements.txt
└── README.md

Data model (star schema)

• Fact — Transactions (weekly grain, already in EUR).
• Dimensions — Products, Customers, Calendar (Jan 2023 → Mar 2025, built in code).
• Budget fact — monthly targets at Country × Category grain.

Critical step: the weekly fact table is aggregated to the monthly grain (Year, Month, Country, Category) so it can be joined to the Budget on (YearMonth, Country, Category). The price/volume variance bridge satisfies the exact identity Price variance + Volume variance = Revenue variance.

Currency: Revenue is already in EUR for all three countries (verified in data_profile.md), so no FX conversion is applied. FX_Rates is loaded for reference only.

Visual standard

• Plotly for every chart, custom template + CSS KPI cards.
• Palette — Teal #0891B2 (actual), Amber #F59E0B (budget), Red #EF4444 (unfavourable), Green #10B981 (favourable).
• Formatting — €#,##0, 0.0%, signed variances; data labels suppressed on charts with >10 points; persistent "Data as of: December 2024" badge.
• Data is cached with @st.cache_data.

Notes

• The Segment slicer does not apply on the Budget Variance page — Budget has no segment dimension (Country × Category × Month only). The page flags this.
• Forecast periods (Jan–Mar 2025) carry no actuals and are modelled from 2024 seasonality times the growth assumption.