Challenge
Leadership data lived in disconnected systems with inconsistent definitions and was reconciled manually each month.
Executive Information System at Alpro
One trusted view of financial, logistics and commercial performance for leadership.
Cloud and on-premise databases, formatted Excel and CSV files harmonised through an automated pipeline with business-owned master data, served in Power BI.
Delivered
A tailored EIS dashboard on an automated pipeline: staging, aggregation, MDS enrichment and DWH loading.
Enabled
Consistent KPIs across departments, with an auditable refresh log.
Challenge
Alpro's Leadership Team needed a single, trusted view of company performance - across financial results, logistics operations, and commercial activity. The underlying data lived in disconnected systems with inconsistent definitions: cloud databases, on-premise databases, manually-maintained Excel files with rich formatting, and CSV exports. Pulling a coherent monthly view for executive decisions meant manual reconciliation across departments - slow, error-prone, and expensive to repeat each cycle.
Approach
We designed a tailored EIS dashboard that aggregates and visualizes the essential KPIs in a single, intuitive interface. The platform ingests data from cloud and on-premise databases, processes formatted Excel and CSV files through Python, and harmonizes everything through a structured pipeline - staging, aggregation, master-data enrichment, and DWH loading - before serving the data to Power BI.
The pipeline is end-to-end automated. Data flows from sources through a Python pre-processing step (handling Excel files with non-trivial formatting, converting to CSV) into a Linux-based filtering pipeline, then into a staging schema for shape normalization. From staging it is aggregated to the month level, populates intermediate fact tables, and lands in MD datamarts.
A Master Data Service (MDS) layer lets business users enrich the dataset with the inputs only they own - currency rates, units, conversion factors. After user enrichment, the dataflow continues automatically: data marts are calculated and loaded from fact tables and the MDS user data, with a load log capturing every refresh and any errors with their reasons. Power BI datasets refresh on top of this trusted layer.
Inside the EIS dashboard
Two screens from the dashboard; figures are blurred.
Dashboard screen · 01 / 02
Leadership overview
Net sales and volume, YTD net sales against target, net sales evolution by quarter and market share by country.
- Month and YTD views
- Net sales against target
- Market share by country
Dashboard screen · 02 / 02
Operations and supply view
Quality losses, operational efficiency, KPI output, raw and pack costs, customer service level by category and sustainability indicators.
- Quality and efficiency by site
- Raw and pack cost trends
- Customer service level by category
Outcomes
Qualitative project outcomes; no measured savings are claimed.
One trusted view of financial, logistical, and commercial performance for the Leadership Team
Real-time visibility into business performance - replacing spreadsheet reconciliation with a governed pipeline
Trend identification and proactive decision-making, not retrospective reporting
Data consistency across departments thanks to a single harmonization layer
Business-owned master data via MDS - currency rates, units, conversion factors maintained where the knowledge lives
Auditable refresh log with explicit error reasons, supporting reliable monthly close
Technology
Python
Linux pipeline
DWH (staging → fact → datamart)
Master Data Services (MDS)
Power BI
Solution areas: Data Foundations · Data Products
