11 brands, one analytics backbone
Designed and built eleven scalable analytical data warehouses for a multi-brand retail franchising group — ten brand-specific DWHs plus a consolidated reporting layer. The architecture lets the holding company buy or divest brands without disrupting analytics, with 21 source systems unified and 80+ reports delivered.
- Client
- International multi-brand retail franchising group
- Industry
- Franchising
- Service
- Data Warehousing
- Engagement
- Dec 2023 — ongoing
What wasn’t working.
The client is a holding company that owns a portfolio of ten retail franchise brands, each operating with its own source systems and data environment. The result was a fragmented analytics estate — no single version of truth across brands, inconsistent definitions of core business metrics, and reporting that couldn't scale with the group's M&A activity.
Three constraints shaped the brief: portfolio flexibility (adding or divesting brands without disrupting analytics), cost discipline (infrastructure cost per brand had to stay lean), and data security (brand-level isolation with consolidated group-level reporting).
How we shipped it.
We designed a federated data warehouse architecture: one DWH per brand for brand-level analytics, plus an independent consolidated DWH for group-level reporting. This gave each brand a clean, isolated analytical layer while the holding company retained a unified view across the portfolio.
The pattern was deliberate — adding an eleventh or twelfth brand is a repeatable implementation, not a re-architecture. Divesting a brand means decommissioning its DWH cleanly without touching the others.
In the box at go-live.
- Eleven analytical data warehouses — ten brand-specific DWHs and one consolidated reporting DWH for group-level analytics.
- 21 source systems integrated across the portfolio, from POS and ERP to franchise management and operational platforms.
- 80+ production reports delivered across brand and group levels, covering sales, operations, franchise performance, and financial reporting.
- Azure Synapse + Microsoft Fabric architecture with Python-based ingestion, Microsoft SQL Server for transactional sources, Power Automate for orchestration, and Power BI as the reporting layer.
- Repeatable brand-onboarding pattern — new acquisitions can be added to the analytical estate without disrupting existing brands.
The number.
A unified analytical platform with consistent metric definitions across the portfolio. A single version of truth established at both brand and holding-company level. The group can now execute M&A without analytics becoming a blocker.
Keep reading.
30 teams on one data estate
Designed and delivered an automated, governed data pipeline for a global development and humanitarian organization — consolidating fragmented Excel-based workflows across multiple functional sections and regions into a centralized Microsoft Fabric warehouse with row-level security and full data lineage.
Three years as data partner to a global retailer
Designed, built, and maintained a production data warehouse for a global retail chain over a three-year direct engagement. Consolidated fragmented reporting into a single source of truth, automated report delivery, and introduced new analytical reports that reshaped how the business made decisions.
What’s the number you want to move?
Thirty minutes with a senior consultant. No pitch deck.